id
stringlengths 9
16
| submitter
stringlengths 3
64
⌀ | authors
stringlengths 5
6.63k
| title
stringlengths 7
245
| comments
stringlengths 1
482
⌀ | journal-ref
stringlengths 4
382
⌀ | doi
stringlengths 9
151
⌀ | report-no
stringclasses 984
values | categories
stringlengths 5
108
| license
stringclasses 9
values | abstract
stringlengths 83
3.41k
| versions
listlengths 1
20
| update_date
timestamp[s]date 2007-05-23 00:00:00
2025-04-11 00:00:00
| authors_parsed
listlengths 1
427
| prompt
stringlengths 166
3.49k
| label
stringclasses 2
values | prob
float64 0.5
0.98
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2503.08207 | Denan Li | Denan Li, Jiyuan Yang, Xiangkai Chen, Lintao Yu, Shi Liu | To Use or Not to Use a Universal Force Field | 21 pages, 5 figures | null | null | null | physics.comp-ph cond-mat.mtrl-sci cs.LG | http://creativecommons.org/licenses/by/4.0/ | Artificial intelligence (AI) is revolutionizing scientific research,
particularly in computational materials science, by enabling more accurate and
efficient simulations. Machine learning force fields (MLFFs) have emerged as
powerful tools for molecular dynamics (MD) simulations, potentially offering
quantum-mechanical accuracy with the efficiency of classical MD. This
Perspective evaluates the viability of universal MLFFs for simulating complex
materials systems from the standpoint of a potential practitioner. Using the
temperature-driven ferroelectric-paraelectric phase transition of PbTiO$_3$ as
a benchmark, we assess leading universal force fields, including CHGNet, MACE,
M3GNet, and GPTFF, alongside specialized models like UniPero. While universal
MLFFs trained on PBE-derived datasets perform well in predicting equilibrium
properties, they largely fail to capture realistic finite-temperature phase
transitions under constant-pressure MD, often exhibiting unphysical
instabilities. These shortcomings stem from inherited biases in
exchange-correlation functionals and limited generalization to anharmonic
interactions governing dynamic behavior. However, fine-tuning universal models
or employing system-specific MLFFs like UniPero successfully restores
predictive accuracy. We advocates for hybrid approaches combining universal
pretraining with targeted optimization, improved error quantification
frameworks, and community-driven benchmarks to advance MLFFs as robust tools
for computational materials discovery.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 09:23:01 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Li",
"Denan",
""
],
[
"Yang",
"Jiyuan",
""
],
[
"Chen",
"Xiangkai",
""
],
[
"Yu",
"Lintao",
""
],
[
"Liu",
"Shi",
""
]
]
| TITLE: To Use or Not to Use a Universal Force Field
ABSTRACT: Artificial intelligence (AI) is revolutionizing scientific research,
particularly in computational materials science, by enabling more accurate and
efficient simulations. Machine learning force fields (MLFFs) have emerged as
powerful tools for molecular dynamics (MD) simulations, potentially offering
quantum-mechanical accuracy with the efficiency of classical MD. This
Perspective evaluates the viability of universal MLFFs for simulating complex
materials systems from the standpoint of a potential practitioner. Using the
temperature-driven ferroelectric-paraelectric phase transition of PbTiO$_3$ as
a benchmark, we assess leading universal force fields, including CHGNet, MACE,
M3GNet, and GPTFF, alongside specialized models like UniPero. While universal
MLFFs trained on PBE-derived datasets perform well in predicting equilibrium
properties, they largely fail to capture realistic finite-temperature phase
transitions under constant-pressure MD, often exhibiting unphysical
instabilities. These shortcomings stem from inherited biases in
exchange-correlation functionals and limited generalization to anharmonic
interactions governing dynamic behavior. However, fine-tuning universal models
or employing system-specific MLFFs like UniPero successfully restores
predictive accuracy. We advocates for hybrid approaches combining universal
pretraining with targeted optimization, improved error quantification
frameworks, and community-driven benchmarks to advance MLFFs as robust tools
for computational materials discovery.
| no_new_dataset | 0.944791 |
2503.08217 | Guangting Zheng | Guangting Zheng, Jiajun Deng, Xiaomeng Chu, Yu Yuan, Houqiang Li and
Yanyong Zhang | S3R-GS: Streamlining the Pipeline for Large-Scale Street Scene
Reconstruction | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, 3D Gaussian Splatting (3DGS) has reshaped the field of
photorealistic 3D reconstruction, achieving impressive rendering quality and
speed. However, when applied to large-scale street scenes, existing methods
suffer from rapidly escalating per-viewpoint reconstruction costs as scene size
increases, leading to significant computational overhead. After revisiting the
conventional pipeline, we identify three key factors accounting for this issue:
unnecessary local-to-global transformations, excessive 3D-to-2D projections,
and inefficient rendering of distant content. To address these challenges, we
propose S3R-GS, a 3DGS framework that Streamlines the pipeline for large-scale
Street Scene Reconstruction, effectively mitigating these limitations.
Moreover, most existing street 3DGS methods rely on ground-truth 3D bounding
boxes to separate dynamic and static components, but 3D bounding boxes are
difficult to obtain, limiting real-world applicability. To address this, we
propose an alternative solution with 2D boxes, which are easier to annotate or
can be predicted by off-the-shelf vision foundation models. Such designs
together make S3R-GS readily adapt to large, in-the-wild scenarios. Extensive
experiments demonstrate that S3R-GS enhances rendering quality and
significantly accelerates reconstruction. Remarkably, when applied to videos
from the challenging Argoverse2 dataset, it achieves state-of-the-art PSNR and
SSIM, reducing reconstruction time to below 50%--and even 20%--of competing
methods.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 09:37:13 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Zheng",
"Guangting",
""
],
[
"Deng",
"Jiajun",
""
],
[
"Chu",
"Xiaomeng",
""
],
[
"Yuan",
"Yu",
""
],
[
"Li",
"Houqiang",
""
],
[
"Zhang",
"Yanyong",
""
]
]
| TITLE: S3R-GS: Streamlining the Pipeline for Large-Scale Street Scene
Reconstruction
ABSTRACT: Recently, 3D Gaussian Splatting (3DGS) has reshaped the field of
photorealistic 3D reconstruction, achieving impressive rendering quality and
speed. However, when applied to large-scale street scenes, existing methods
suffer from rapidly escalating per-viewpoint reconstruction costs as scene size
increases, leading to significant computational overhead. After revisiting the
conventional pipeline, we identify three key factors accounting for this issue:
unnecessary local-to-global transformations, excessive 3D-to-2D projections,
and inefficient rendering of distant content. To address these challenges, we
propose S3R-GS, a 3DGS framework that Streamlines the pipeline for large-scale
Street Scene Reconstruction, effectively mitigating these limitations.
Moreover, most existing street 3DGS methods rely on ground-truth 3D bounding
boxes to separate dynamic and static components, but 3D bounding boxes are
difficult to obtain, limiting real-world applicability. To address this, we
propose an alternative solution with 2D boxes, which are easier to annotate or
can be predicted by off-the-shelf vision foundation models. Such designs
together make S3R-GS readily adapt to large, in-the-wild scenarios. Extensive
experiments demonstrate that S3R-GS enhances rendering quality and
significantly accelerates reconstruction. Remarkably, when applied to videos
from the challenging Argoverse2 dataset, it achieves state-of-the-art PSNR and
SSIM, reducing reconstruction time to below 50%--and even 20%--of competing
methods.
| no_new_dataset | 0.942348 |
2503.08218 | Kaiqiang Xiong | Kaiqiang Xiong, Ying Feng, Qi Zhang, Jianbo Jiao, Yang Zhao, Zhihao
Liang, Huachen Gao, Ronggang Wang | MVD-HuGaS: Human Gaussians from a Single Image via 3D Human Multi-view
Diffusion Prior | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | 3D human reconstruction from a single image is a challenging problem and has
been exclusively studied in the literature. Recently, some methods have
resorted to diffusion models for guidance, optimizing a 3D representation via
Score Distillation Sampling(SDS) or generating one back-view image for
facilitating reconstruction. However, these methods tend to produce
unsatisfactory artifacts (\textit{e.g.} flattened human structure or
over-smoothing results caused by inconsistent priors from multiple views) and
struggle with real-world generalization in the wild. In this work, we present
\emph{MVD-HuGaS}, enabling free-view 3D human rendering from a single image via
a multi-view human diffusion model. We first generate multi-view images from
the single reference image with an enhanced multi-view diffusion model, which
is well fine-tuned on high-quality 3D human datasets to incorporate 3D geometry
priors and human structure priors. To infer accurate camera poses from the
sparse generated multi-view images for reconstruction, an alignment module is
introduced to facilitate joint optimization of 3D Gaussians and camera poses.
Furthermore, we propose a depth-based Facial Distortion Mitigation module to
refine the generated facial regions, thereby improving the overall fidelity of
the reconstruction.Finally, leveraging the refined multi-view images, along
with their accurate camera poses, MVD-HuGaS optimizes the 3D Gaussians of the
target human for high-fidelity free-view renderings. Extensive experiments on
Thuman2.0 and 2K2K datasets show that the proposed MVD-HuGaS achieves
state-of-the-art performance on single-view 3D human rendering.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 09:37:15 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Xiong",
"Kaiqiang",
""
],
[
"Feng",
"Ying",
""
],
[
"Zhang",
"Qi",
""
],
[
"Jiao",
"Jianbo",
""
],
[
"Zhao",
"Yang",
""
],
[
"Liang",
"Zhihao",
""
],
[
"Gao",
"Huachen",
""
],
[
"Wang",
"Ronggang",
""
]
]
| TITLE: MVD-HuGaS: Human Gaussians from a Single Image via 3D Human Multi-view
Diffusion Prior
ABSTRACT: 3D human reconstruction from a single image is a challenging problem and has
been exclusively studied in the literature. Recently, some methods have
resorted to diffusion models for guidance, optimizing a 3D representation via
Score Distillation Sampling(SDS) or generating one back-view image for
facilitating reconstruction. However, these methods tend to produce
unsatisfactory artifacts (\textit{e.g.} flattened human structure or
over-smoothing results caused by inconsistent priors from multiple views) and
struggle with real-world generalization in the wild. In this work, we present
\emph{MVD-HuGaS}, enabling free-view 3D human rendering from a single image via
a multi-view human diffusion model. We first generate multi-view images from
the single reference image with an enhanced multi-view diffusion model, which
is well fine-tuned on high-quality 3D human datasets to incorporate 3D geometry
priors and human structure priors. To infer accurate camera poses from the
sparse generated multi-view images for reconstruction, an alignment module is
introduced to facilitate joint optimization of 3D Gaussians and camera poses.
Furthermore, we propose a depth-based Facial Distortion Mitigation module to
refine the generated facial regions, thereby improving the overall fidelity of
the reconstruction.Finally, leveraging the refined multi-view images, along
with their accurate camera poses, MVD-HuGaS optimizes the 3D Gaussians of the
target human for high-fidelity free-view renderings. Extensive experiments on
Thuman2.0 and 2K2K datasets show that the proposed MVD-HuGaS achieves
state-of-the-art performance on single-view 3D human rendering.
| no_new_dataset | 0.9549 |
2503.08221 | Junbin Xiao | Junbin Xiao, Nanxin Huang, Hao Qiu, Zhulin Tao, Xun Yang, Richang
Hong, Meng Wang, Angela Yao | EgoBlind: Towards Egocentric Visual Assistance for the Blind People | Preprint. Under Review | null | null | null | cs.CV cs.AI cs.MM | http://creativecommons.org/licenses/by-nc-sa/4.0/ | We present EgoBlind, the first egocentric VideoQA dataset collected from
blind individuals to evaluate the assistive capabilities of contemporary
multimodal large language models (MLLMs). EgoBlind comprises 1,210 videos that
record the daily lives of real blind users from a first-person perspective. It
also features 4,927 questions directly posed or generated and verified by blind
individuals to reflect their needs for visual assistance under various
scenarios. We provide each question with an average of 3 reference answers to
alleviate subjective evaluation. Using EgoBlind, we comprehensively evaluate 15
leading MLLMs and find that all models struggle, with the best performers
achieving accuracy around 56\%, far behind human performance of 87.4\%. To
guide future advancements, we identify and summarize major limitations of
existing MLLMs in egocentric visual assistance for the blind and provide
heuristic suggestions for improvement. With these efforts, we hope EgoBlind can
serve as a valuable foundation for developing more effective AI assistants to
enhance the independence of the blind individuals' lives.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 09:40:31 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Xiao",
"Junbin",
""
],
[
"Huang",
"Nanxin",
""
],
[
"Qiu",
"Hao",
""
],
[
"Tao",
"Zhulin",
""
],
[
"Yang",
"Xun",
""
],
[
"Hong",
"Richang",
""
],
[
"Wang",
"Meng",
""
],
[
"Yao",
"Angela",
""
]
]
| TITLE: EgoBlind: Towards Egocentric Visual Assistance for the Blind People
ABSTRACT: We present EgoBlind, the first egocentric VideoQA dataset collected from
blind individuals to evaluate the assistive capabilities of contemporary
multimodal large language models (MLLMs). EgoBlind comprises 1,210 videos that
record the daily lives of real blind users from a first-person perspective. It
also features 4,927 questions directly posed or generated and verified by blind
individuals to reflect their needs for visual assistance under various
scenarios. We provide each question with an average of 3 reference answers to
alleviate subjective evaluation. Using EgoBlind, we comprehensively evaluate 15
leading MLLMs and find that all models struggle, with the best performers
achieving accuracy around 56\%, far behind human performance of 87.4\%. To
guide future advancements, we identify and summarize major limitations of
existing MLLMs in egocentric visual assistance for the blind and provide
heuristic suggestions for improvement. With these efforts, we hope EgoBlind can
serve as a valuable foundation for developing more effective AI assistants to
enhance the independence of the blind individuals' lives.
| new_dataset | 0.961316 |
2503.08239 | Muhammad Ahmad | Saad Sohail, Muhammad Usama, Usman Ghous, Manuel Mazzara, Salvatore
Distefano, Muhammad Ahmad | EnergyFormer: Energy Attention with Fourier Embedding for Hyperspectral
Image Classification | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Hyperspectral imaging (HSI) provides rich spectral-spatial information across
hundreds of contiguous bands, enabling precise material discrimination in
applications such as environmental monitoring, agriculture, and urban analysis.
However, the high dimensionality and spectral variability of HSI data pose
significant challenges for feature extraction and classification. This paper
presents EnergyFormer, a transformer-based framework designed to address these
challenges through three key innovations: (1) Multi-Head Energy Attention
(MHEA), which optimizes an energy function to selectively enhance critical
spectral-spatial features, improving feature discrimination; (2) Fourier
Position Embedding (FoPE), which adaptively encodes spectral and spatial
dependencies to reinforce long-range interactions; and (3) Enhanced
Convolutional Block Attention Module (ECBAM), which selectively amplifies
informative wavelength bands and spatial structures, enhancing representation
learning. Extensive experiments on the WHU-Hi-HanChuan, Salinas, and Pavia
University datasets demonstrate that EnergyFormer achieves exceptional overall
accuracies of 99.28\%, 98.63\%, and 98.72\%, respectively, outperforming
state-of-the-art CNN, transformer, and Mamba-based models. The source code will
be made available at https://github.com/mahmad000.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 10:03:35 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Sohail",
"Saad",
""
],
[
"Usama",
"Muhammad",
""
],
[
"Ghous",
"Usman",
""
],
[
"Mazzara",
"Manuel",
""
],
[
"Distefano",
"Salvatore",
""
],
[
"Ahmad",
"Muhammad",
""
]
]
| TITLE: EnergyFormer: Energy Attention with Fourier Embedding for Hyperspectral
Image Classification
ABSTRACT: Hyperspectral imaging (HSI) provides rich spectral-spatial information across
hundreds of contiguous bands, enabling precise material discrimination in
applications such as environmental monitoring, agriculture, and urban analysis.
However, the high dimensionality and spectral variability of HSI data pose
significant challenges for feature extraction and classification. This paper
presents EnergyFormer, a transformer-based framework designed to address these
challenges through three key innovations: (1) Multi-Head Energy Attention
(MHEA), which optimizes an energy function to selectively enhance critical
spectral-spatial features, improving feature discrimination; (2) Fourier
Position Embedding (FoPE), which adaptively encodes spectral and spatial
dependencies to reinforce long-range interactions; and (3) Enhanced
Convolutional Block Attention Module (ECBAM), which selectively amplifies
informative wavelength bands and spatial structures, enhancing representation
learning. Extensive experiments on the WHU-Hi-HanChuan, Salinas, and Pavia
University datasets demonstrate that EnergyFormer achieves exceptional overall
accuracies of 99.28\%, 98.63\%, and 98.72\%, respectively, outperforming
state-of-the-art CNN, transformer, and Mamba-based models. The source code will
be made available at https://github.com/mahmad000.
| no_new_dataset | 0.955817 |
2503.08240 | Lachlan Simpson | Lachlan Simpson, Federico Costanza, Kyle Millar, Adriel Cheng,
Cheng-Chew Lim, Hong Gunn Chew | Tangentially Aligned Integrated Gradients for User-Friendly Explanations | To appear in the proceedings of the 32nd Irish Conference on
Artificial Intelligence and Cognitive Science | null | null | null | cs.LG math.DG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Integrated gradients is prevalent within machine learning to address the
black-box problem of neural networks. The explanations given by integrated
gradients depend on a choice of base-point. The choice of base-point is not a
priori obvious and can lead to drastically different explanations. There is a
longstanding hypothesis that data lies on a low dimensional Riemannian
manifold. The quality of explanations on a manifold can be measured by the
extent to which an explanation for a point lies in its tangent space. In this
work, we propose that the base-point should be chosen such that it maximises
the tangential alignment of the explanation. We formalise the notion of
tangential alignment and provide theoretical conditions under which a
base-point choice will provide explanations lying in the tangent space. We
demonstrate how to approximate the optimal base-point on several well-known
image classification datasets. Furthermore, we compare the optimal base-point
choice with common base-points and three gradient explainability models.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 10:04:13 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Simpson",
"Lachlan",
""
],
[
"Costanza",
"Federico",
""
],
[
"Millar",
"Kyle",
""
],
[
"Cheng",
"Adriel",
""
],
[
"Lim",
"Cheng-Chew",
""
],
[
"Chew",
"Hong Gunn",
""
]
]
| TITLE: Tangentially Aligned Integrated Gradients for User-Friendly Explanations
ABSTRACT: Integrated gradients is prevalent within machine learning to address the
black-box problem of neural networks. The explanations given by integrated
gradients depend on a choice of base-point. The choice of base-point is not a
priori obvious and can lead to drastically different explanations. There is a
longstanding hypothesis that data lies on a low dimensional Riemannian
manifold. The quality of explanations on a manifold can be measured by the
extent to which an explanation for a point lies in its tangent space. In this
work, we propose that the base-point should be chosen such that it maximises
the tangential alignment of the explanation. We formalise the notion of
tangential alignment and provide theoretical conditions under which a
base-point choice will provide explanations lying in the tangent space. We
demonstrate how to approximate the optimal base-point on several well-known
image classification datasets. Furthermore, we compare the optimal base-point
choice with common base-points and three gradient explainability models.
| no_new_dataset | 0.951953 |
2503.08246 | Peter Macgregor | Seiyun Shin, Ilan Shomorony, Peter Macgregor | Dynamic DBSCAN with Euler Tour Sequences | AISTATS 2025 | null | null | null | cs.LG cs.DS | http://creativecommons.org/licenses/by/4.0/ | We propose a fast and dynamic algorithm for Density-Based Spatial Clustering
of Applications with Noise (DBSCAN) that efficiently supports online updates.
Traditional DBSCAN algorithms, designed for batch processing, become
computationally expensive when applied to dynamic datasets, particularly in
large-scale applications where data continuously evolves. To address this
challenge, our algorithm leverages the Euler Tour Trees data structure,
enabling dynamic clustering updates without the need to reprocess the entire
dataset. This approach preserves a near-optimal accuracy in density estimation,
as achieved by the state-of-the-art static DBSCAN method (Esfandiari et al.,
2021) Our method achieves an improved time complexity of $O(d \log^3(n) +
\log^4(n))$ for every data point insertion and deletion, where $n$ and $d$
denote the total number of updates and the data dimension, respectively.
Empirical studies also demonstrate significant speedups over conventional
DBSCANs in real-time clustering of dynamic datasets, while maintaining
comparable or superior clustering quality.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 10:08:39 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Shin",
"Seiyun",
""
],
[
"Shomorony",
"Ilan",
""
],
[
"Macgregor",
"Peter",
""
]
]
| TITLE: Dynamic DBSCAN with Euler Tour Sequences
ABSTRACT: We propose a fast and dynamic algorithm for Density-Based Spatial Clustering
of Applications with Noise (DBSCAN) that efficiently supports online updates.
Traditional DBSCAN algorithms, designed for batch processing, become
computationally expensive when applied to dynamic datasets, particularly in
large-scale applications where data continuously evolves. To address this
challenge, our algorithm leverages the Euler Tour Trees data structure,
enabling dynamic clustering updates without the need to reprocess the entire
dataset. This approach preserves a near-optimal accuracy in density estimation,
as achieved by the state-of-the-art static DBSCAN method (Esfandiari et al.,
2021) Our method achieves an improved time complexity of $O(d \log^3(n) +
\log^4(n))$ for every data point insertion and deletion, where $n$ and $d$
denote the total number of updates and the data dimension, respectively.
Empirical studies also demonstrate significant speedups over conventional
DBSCANs in real-time clustering of dynamic datasets, while maintaining
comparable or superior clustering quality.
| no_new_dataset | 0.94699 |
2503.08251 | Arshia Afzal | Arshia Afzal and Volkan Cevher and Mahsa Shoaran | MT-NAM: An Efficient and Adaptive Model for Epileptic Seizure Detection | Submitted to IEEE-TBME | null | null | null | eess.SP cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | Enhancing the accuracy and efficiency of machine learning algorithms employed
in neural interface systems is crucial for advancing next-generation
intelligent therapeutic devices. However, current systems often utilize basic
machine learning models that do not fully exploit the natural structure of
brain signals. Additionally, existing learning models used for neural signal
processing often demonstrate low speed and efficiency during inference. To
address these challenges, this study introduces Micro Tree-based NAM (MT-NAM),
a distilled model based on the recently proposed Neural Additive Models (NAM).
The MT-NAM achieves a remarkable 100$\times$ improvement in inference speed
compared to standard NAM, without compromising accuracy. We evaluate our
approach on the CHB-MIT scalp EEG dataset, which includes recordings from 24
patients with varying numbers of sessions and seizures. NAM achieves an 85.3\%
window-based sensitivity and 95\% specificity. Interestingly, our proposed
MT-NAM shows only a 2\% reduction in sensitivity compared to the original NAM.
To regain this sensitivity, we utilize a test-time template adjuster (T3A) as
an update mechanism, enabling our model to achieve higher sensitivity during
test time by accommodating transient shifts in neural signals. With this online
update approach, MT-NAM achieves the same sensitivity as the standard NAM while
achieving approximately 50$\times$ acceleration in inference speed.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 10:14:53 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Afzal",
"Arshia",
""
],
[
"Cevher",
"Volkan",
""
],
[
"Shoaran",
"Mahsa",
""
]
]
| TITLE: MT-NAM: An Efficient and Adaptive Model for Epileptic Seizure Detection
ABSTRACT: Enhancing the accuracy and efficiency of machine learning algorithms employed
in neural interface systems is crucial for advancing next-generation
intelligent therapeutic devices. However, current systems often utilize basic
machine learning models that do not fully exploit the natural structure of
brain signals. Additionally, existing learning models used for neural signal
processing often demonstrate low speed and efficiency during inference. To
address these challenges, this study introduces Micro Tree-based NAM (MT-NAM),
a distilled model based on the recently proposed Neural Additive Models (NAM).
The MT-NAM achieves a remarkable 100$\times$ improvement in inference speed
compared to standard NAM, without compromising accuracy. We evaluate our
approach on the CHB-MIT scalp EEG dataset, which includes recordings from 24
patients with varying numbers of sessions and seizures. NAM achieves an 85.3\%
window-based sensitivity and 95\% specificity. Interestingly, our proposed
MT-NAM shows only a 2\% reduction in sensitivity compared to the original NAM.
To regain this sensitivity, we utilize a test-time template adjuster (T3A) as
an update mechanism, enabling our model to achieve higher sensitivity during
test time by accommodating transient shifts in neural signals. With this online
update approach, MT-NAM achieves the same sensitivity as the standard NAM while
achieving approximately 50$\times$ acceleration in inference speed.
| no_new_dataset | 0.946349 |
2503.08270 | Chengjun Yu | Chengjun Yu, Wei Zhai, Yuhang Yang, Yang Cao, Zheng-Jun Zha | HERO: Human Reaction Generation from Videos | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Human reaction generation represents a significant research domain for
interactive AI, as humans constantly interact with their surroundings. Previous
works focus mainly on synthesizing the reactive motion given a human motion
sequence. This paradigm limits interaction categories to human-human
interactions and ignores emotions that may influence reaction generation. In
this work, we propose to generate 3D human reactions from RGB videos, which
involves a wider range of interaction categories and naturally provides
information about expressions that may reflect the subject's emotions. To cope
with this task, we present HERO, a simple yet powerful framework for Human
rEaction geneRation from videOs. HERO considers both global and frame-level
local representations of the video to extract the interaction intention, and
then uses the extracted interaction intention to guide the synthesis of the
reaction. Besides, local visual representations are continuously injected into
the model to maximize the exploitation of the dynamic properties inherent in
videos. Furthermore, the ViMo dataset containing paired Video-Motion data is
collected to support the task. In addition to human-human interactions, these
video-motion pairs also cover animal-human interactions and scene-human
interactions. Extensive experiments demonstrate the superiority of our
methodology. The code and dataset will be publicly available at
https://jackyu6.github.io/HERO.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 10:39:32 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Yu",
"Chengjun",
""
],
[
"Zhai",
"Wei",
""
],
[
"Yang",
"Yuhang",
""
],
[
"Cao",
"Yang",
""
],
[
"Zha",
"Zheng-Jun",
""
]
]
| TITLE: HERO: Human Reaction Generation from Videos
ABSTRACT: Human reaction generation represents a significant research domain for
interactive AI, as humans constantly interact with their surroundings. Previous
works focus mainly on synthesizing the reactive motion given a human motion
sequence. This paradigm limits interaction categories to human-human
interactions and ignores emotions that may influence reaction generation. In
this work, we propose to generate 3D human reactions from RGB videos, which
involves a wider range of interaction categories and naturally provides
information about expressions that may reflect the subject's emotions. To cope
with this task, we present HERO, a simple yet powerful framework for Human
rEaction geneRation from videOs. HERO considers both global and frame-level
local representations of the video to extract the interaction intention, and
then uses the extracted interaction intention to guide the synthesis of the
reaction. Besides, local visual representations are continuously injected into
the model to maximize the exploitation of the dynamic properties inherent in
videos. Furthermore, the ViMo dataset containing paired Video-Motion data is
collected to support the task. In addition to human-human interactions, these
video-motion pairs also cover animal-human interactions and scene-human
interactions. Extensive experiments demonstrate the superiority of our
methodology. The code and dataset will be publicly available at
https://jackyu6.github.io/HERO.
| no_new_dataset | 0.936981 |
2503.08271 | Wenzhe Niu | Wenzhe Niu, Zongxia Xie, Yanru Sun, Wei He, Man Xu, Chao Hao | LangTime: A Language-Guided Unified Model for Time Series Forecasting
with Proximal Policy Optimization | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Recent research has shown an increasing interest in utilizing pre-trained
large language models (LLMs) for a variety of time series applications.
However, there are three main challenges when using LLMs as foundational models
for time series forecasting: (1) Cross-domain generalization. (2)
Cross-modality alignment. (3) Error accumulation in autoregressive frameworks.
To address these challenges, we proposed LangTime, a language-guided unified
model for time series forecasting that incorporates cross-domain pre-training
with reinforcement learning-based fine-tuning. Specifically, LangTime
constructs Temporal Comprehension Prompts (TCPs), which include dataset-wise
and channel-wise instructions, to facilitate domain adaptation and condense
time series into a single token, enabling LLMs to understand better and align
temporal data. To improve autoregressive forecasting, we introduce TimePPO, a
reinforcement learning-based fine-tuning algorithm. TimePPO mitigates error
accumulation by leveraging a multidimensional rewards function tailored for
time series and a repeat-based value estimation strategy. Extensive experiments
demonstrate that LangTime achieves state-of-the-art cross-domain forecasting
performance, while TimePPO fine-tuning effectively enhances the stability and
accuracy of autoregressive forecasting.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 10:40:39 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Niu",
"Wenzhe",
""
],
[
"Xie",
"Zongxia",
""
],
[
"Sun",
"Yanru",
""
],
[
"He",
"Wei",
""
],
[
"Xu",
"Man",
""
],
[
"Hao",
"Chao",
""
]
]
| TITLE: LangTime: A Language-Guided Unified Model for Time Series Forecasting
with Proximal Policy Optimization
ABSTRACT: Recent research has shown an increasing interest in utilizing pre-trained
large language models (LLMs) for a variety of time series applications.
However, there are three main challenges when using LLMs as foundational models
for time series forecasting: (1) Cross-domain generalization. (2)
Cross-modality alignment. (3) Error accumulation in autoregressive frameworks.
To address these challenges, we proposed LangTime, a language-guided unified
model for time series forecasting that incorporates cross-domain pre-training
with reinforcement learning-based fine-tuning. Specifically, LangTime
constructs Temporal Comprehension Prompts (TCPs), which include dataset-wise
and channel-wise instructions, to facilitate domain adaptation and condense
time series into a single token, enabling LLMs to understand better and align
temporal data. To improve autoregressive forecasting, we introduce TimePPO, a
reinforcement learning-based fine-tuning algorithm. TimePPO mitigates error
accumulation by leveraging a multidimensional rewards function tailored for
time series and a repeat-based value estimation strategy. Extensive experiments
demonstrate that LangTime achieves state-of-the-art cross-domain forecasting
performance, while TimePPO fine-tuning effectively enhances the stability and
accuracy of autoregressive forecasting.
| no_new_dataset | 0.947478 |
2503.08276 | Miao Zhang | Jun Yin, Yangfan He, Miao Zhang, Pengyu Zeng, Tianyi Wang, Shuai Lu,
Xueqian Wang | PromptLNet: Region-Adaptive Aesthetic Enhancement via Prompt Guidance in
Low-Light Enhancement Net | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Learning and improving large language models through human preference
feedback has become a mainstream approach, but it has rarely been applied to
the field of low-light image enhancement. Existing low-light enhancement
evaluations typically rely on objective metrics (such as FID, PSNR, etc.),
which often result in models that perform well objectively but lack aesthetic
quality. Moreover, most low-light enhancement models are primarily designed for
global brightening, lacking detailed refinement. Therefore, the generated
images often require additional local adjustments, leading to research gaps in
practical applications. To bridge this gap, we propose the following
innovations: 1) We collect human aesthetic evaluation text pairs and aesthetic
scores from multiple low-light image datasets (e.g., LOL, LOL2, LOM, DCIM, MEF,
etc.) to train a low-light image aesthetic evaluation model, supplemented by an
optimization algorithm designed to fine-tune the diffusion model. 2) We propose
a prompt-driven brightness adjustment module capable of performing fine-grained
brightness and aesthetic adjustments for specific instances or regions. 3) We
evaluate our method alongside existing state-of-the-art algorithms on
mainstream benchmarks. Experimental results show that our method not only
outperforms traditional methods in terms of visual quality but also provides
greater flexibility and controllability, paving the way for improved aesthetic
quality.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 10:45:08 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Yin",
"Jun",
""
],
[
"He",
"Yangfan",
""
],
[
"Zhang",
"Miao",
""
],
[
"Zeng",
"Pengyu",
""
],
[
"Wang",
"Tianyi",
""
],
[
"Lu",
"Shuai",
""
],
[
"Wang",
"Xueqian",
""
]
]
| TITLE: PromptLNet: Region-Adaptive Aesthetic Enhancement via Prompt Guidance in
Low-Light Enhancement Net
ABSTRACT: Learning and improving large language models through human preference
feedback has become a mainstream approach, but it has rarely been applied to
the field of low-light image enhancement. Existing low-light enhancement
evaluations typically rely on objective metrics (such as FID, PSNR, etc.),
which often result in models that perform well objectively but lack aesthetic
quality. Moreover, most low-light enhancement models are primarily designed for
global brightening, lacking detailed refinement. Therefore, the generated
images often require additional local adjustments, leading to research gaps in
practical applications. To bridge this gap, we propose the following
innovations: 1) We collect human aesthetic evaluation text pairs and aesthetic
scores from multiple low-light image datasets (e.g., LOL, LOL2, LOM, DCIM, MEF,
etc.) to train a low-light image aesthetic evaluation model, supplemented by an
optimization algorithm designed to fine-tune the diffusion model. 2) We propose
a prompt-driven brightness adjustment module capable of performing fine-grained
brightness and aesthetic adjustments for specific instances or regions. 3) We
evaluate our method alongside existing state-of-the-art algorithms on
mainstream benchmarks. Experimental results show that our method not only
outperforms traditional methods in terms of visual quality but also provides
greater flexibility and controllability, paving the way for improved aesthetic
quality.
| no_new_dataset | 0.950041 |
2503.08290 | Sachin Verma | Sachin Verma, Frank Lindseth, Gabriel Kiss | SegDesicNet: Lightweight Semantic Segmentation in Remote Sensing with
Geo-Coordinate Embeddings for Domain Adaptation | https://openaccess.thecvf.com/content/WACV2025/papers/Verma_SegDesicNet_Lightweight_Semantic_Segmentation_in_Remote_Sensing_with_Geo-Coordinate_Embeddings_WACV_2025_paper.pdf | IEEE/CVF Winter Conference on Applications of Computer Vision
(WACV) 2025 | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Semantic segmentation is essential for analyzing highdefinition remote
sensing images (HRSIs) because it allows the precise classification of objects
and regions at the pixel level. However, remote sensing data present challenges
owing to geographical location, weather, and environmental variations, making
it difficult for semantic segmentation models to generalize across diverse
scenarios. Existing methods are often limited to specific data domains and
require expert annotators and specialized equipment for semantic labeling. In
this study, we propose a novel unsupervised domain adaptation technique for
remote sensing semantic segmentation by utilizing geographical coordinates that
are readily accessible in remote sensing setups as metadata in a dataset. To
bridge the domain gap, we propose a novel approach that considers the
combination of an image\'s location encoding trait and the spherical nature of
Earth\'s surface. Our proposed SegDesicNet module regresses the GRID positional
encoding of the geo coordinates projected over the unit sphere to obtain the
domain loss. Our experimental results demonstrate that the proposed SegDesicNet
outperforms state of the art domain adaptation methods in remote sensing image
segmentation, achieving an improvement of approximately ~6% in the mean
intersection over union (MIoU) with a ~ 27\% drop in parameter count on
benchmarked subsets of the publicly available FLAIR #1 dataset. We also
benchmarked our method performance on the custom split of the ISPRS Potsdam
dataset. Our algorithm seeks to reduce the modeling disparity between
artificial neural networks and human comprehension of the physical world,
making the technology more human centric and scalable.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 11:01:18 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Verma",
"Sachin",
""
],
[
"Lindseth",
"Frank",
""
],
[
"Kiss",
"Gabriel",
""
]
]
| TITLE: SegDesicNet: Lightweight Semantic Segmentation in Remote Sensing with
Geo-Coordinate Embeddings for Domain Adaptation
ABSTRACT: Semantic segmentation is essential for analyzing highdefinition remote
sensing images (HRSIs) because it allows the precise classification of objects
and regions at the pixel level. However, remote sensing data present challenges
owing to geographical location, weather, and environmental variations, making
it difficult for semantic segmentation models to generalize across diverse
scenarios. Existing methods are often limited to specific data domains and
require expert annotators and specialized equipment for semantic labeling. In
this study, we propose a novel unsupervised domain adaptation technique for
remote sensing semantic segmentation by utilizing geographical coordinates that
are readily accessible in remote sensing setups as metadata in a dataset. To
bridge the domain gap, we propose a novel approach that considers the
combination of an image\'s location encoding trait and the spherical nature of
Earth\'s surface. Our proposed SegDesicNet module regresses the GRID positional
encoding of the geo coordinates projected over the unit sphere to obtain the
domain loss. Our experimental results demonstrate that the proposed SegDesicNet
outperforms state of the art domain adaptation methods in remote sensing image
segmentation, achieving an improvement of approximately ~6% in the mean
intersection over union (MIoU) with a ~ 27\% drop in parameter count on
benchmarked subsets of the publicly available FLAIR #1 dataset. We also
benchmarked our method performance on the custom split of the ISPRS Potsdam
dataset. Our algorithm seeks to reduce the modeling disparity between
artificial neural networks and human comprehension of the physical world,
making the technology more human centric and scalable.
| no_new_dataset | 0.9549 |
2503.08293 | Alberto Miguel Diez | Alberto Miguel-Diez, Adri\'an Campazas-Vega, Claudia
\'Alvarez-Aparicio, Gonzalo Esteban-Costales, \'Angel Manuel
Guerrero-Higueras | A systematic literature review of unsupervised learning algorithms for
anomalous traffic detection based on flows | This article has been accepted for publication in Logic Journal of
the IGPL Published by Oxford University Press | null | null | null | cs.CR cs.LG cs.NI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The constant increase of devices connected to the Internet, and therefore of
cyber-attacks, makes it necessary to analyze network traffic in order to
recognize malicious activity. Traditional packet-based analysis methods are
insufficient because in large networks the amount of traffic is so high that it
is unfeasible to review all communications. For this reason, flows is a
suitable approach for this situation, which in future 5G networks will have to
be used, as the number of packets will increase dramatically. If this is also
combined with unsupervised learning models, it can detect new threats for which
it has not been trained. This paper presents a systematic review of the
literature on unsupervised learning algorithms for detecting anomalies in
network flows, following the PRISMA guideline. A total of 63 scientific
articles have been reviewed, analyzing 13 of them in depth. The results
obtained show that autoencoder is the most used option, followed by SVM, ALAD,
or SOM. On the other hand, all the datasets used for anomaly detection have
been collected, including some specialised in IoT or with real data collected
from honeypots.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 11:06:00 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Miguel-Diez",
"Alberto",
""
],
[
"Campazas-Vega",
"Adrián",
""
],
[
"Álvarez-Aparicio",
"Claudia",
""
],
[
"Esteban-Costales",
"Gonzalo",
""
],
[
"Guerrero-Higueras",
"Ángel Manuel",
""
]
]
| TITLE: A systematic literature review of unsupervised learning algorithms for
anomalous traffic detection based on flows
ABSTRACT: The constant increase of devices connected to the Internet, and therefore of
cyber-attacks, makes it necessary to analyze network traffic in order to
recognize malicious activity. Traditional packet-based analysis methods are
insufficient because in large networks the amount of traffic is so high that it
is unfeasible to review all communications. For this reason, flows is a
suitable approach for this situation, which in future 5G networks will have to
be used, as the number of packets will increase dramatically. If this is also
combined with unsupervised learning models, it can detect new threats for which
it has not been trained. This paper presents a systematic review of the
literature on unsupervised learning algorithms for detecting anomalies in
network flows, following the PRISMA guideline. A total of 63 scientific
articles have been reviewed, analyzing 13 of them in depth. The results
obtained show that autoencoder is the most used option, followed by SVM, ALAD,
or SOM. On the other hand, all the datasets used for anomaly detection have
been collected, including some specialised in IoT or with real data collected
from honeypots.
| no_new_dataset | 0.940298 |
2503.08298 | George Papadakis | Jakub Maciejewski, Konstantinos Nikoletos, George Papadakis, Yannis
Velegrakis | Progressive Entity Resolution: A Design Space Exploration | null | null | 10.1145/3709715 | null | cs.DB | http://creativecommons.org/licenses/by/4.0/ | Entity Resolution (ER) is typically implemented as a batch task that
processes all available data before identifying duplicate records. However,
applications with time or computational constraints, e.g., those running in the
cloud, require a progressive approach that produces results in a pay-as-you-go
fashion. Numerous algorithms have been proposed for Progressive ER in the
literature. In this work, we propose a novel framework for Progressive Entity
Resolution that organizes relevant techniques into four consecutive steps: (i)
filtering, which reduces the search space to the most likely candidate matches,
(ii) weighting, which associates every pair of candidate matches with a
similarity score, (iii) scheduling, which prioritizes the execution of the
candidate matches so that the real duplicates precede the non-matching pairs,
and (iv) matching, which applies a complex, matching function to the pairs in
the order defined by the previous step. We associate each step with existing
and novel techniques, illustrating that our framework overall generates a
superset of the main existing works in the field. We select the most
representative combinations resulting from our framework and fine-tune them
over 10 established datasets for Record Linkage and 8 for Deduplication, with
our results indicating that our taxonomy yields a wide range of high performing
progressive techniques both in terms of effectiveness and time efficiency.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 11:10:15 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Maciejewski",
"Jakub",
""
],
[
"Nikoletos",
"Konstantinos",
""
],
[
"Papadakis",
"George",
""
],
[
"Velegrakis",
"Yannis",
""
]
]
| TITLE: Progressive Entity Resolution: A Design Space Exploration
ABSTRACT: Entity Resolution (ER) is typically implemented as a batch task that
processes all available data before identifying duplicate records. However,
applications with time or computational constraints, e.g., those running in the
cloud, require a progressive approach that produces results in a pay-as-you-go
fashion. Numerous algorithms have been proposed for Progressive ER in the
literature. In this work, we propose a novel framework for Progressive Entity
Resolution that organizes relevant techniques into four consecutive steps: (i)
filtering, which reduces the search space to the most likely candidate matches,
(ii) weighting, which associates every pair of candidate matches with a
similarity score, (iii) scheduling, which prioritizes the execution of the
candidate matches so that the real duplicates precede the non-matching pairs,
and (iv) matching, which applies a complex, matching function to the pairs in
the order defined by the previous step. We associate each step with existing
and novel techniques, illustrating that our framework overall generates a
superset of the main existing works in the field. We select the most
representative combinations resulting from our framework and fine-tune them
over 10 established datasets for Record Linkage and 8 for Deduplication, with
our results indicating that our taxonomy yields a wide range of high performing
progressive techniques both in terms of effectiveness and time efficiency.
| no_new_dataset | 0.951908 |
2503.08308 | Zhuo Zhi | Zhuo Zhi, Chen Feng, Adam Daneshmend, Mine Orlu, Andreas Demosthenous,
Lu Yin, Da Li, Ziquan Liu, Miguel R. D. Rodrigues | Seeing and Reasoning with Confidence: Supercharging Multimodal LLMs with
an Uncertainty-Aware Agentic Framework | null | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | Multimodal large language models (MLLMs) show promise in tasks like visual
question answering (VQA) but still face challenges in multimodal reasoning.
Recent works adapt agentic frameworks or chain-of-thought (CoT) reasoning to
improve performance. However, CoT-based multimodal reasoning often demands
costly data annotation and fine-tuning, while agentic approaches relying on
external tools risk introducing unreliable output from these tools. In this
paper, we propose Seeing and Reasoning with Confidence (SRICE), a training-free
multimodal reasoning framework that integrates external vision models with
uncertainty quantification (UQ) into an MLLM to address these challenges.
Specifically, SRICE guides the inference process by allowing MLLM to
autonomously select regions of interest through multi-stage interactions with
the help of external tools. We propose to use a conformal prediction-based
approach to calibrate the output of external tools and select the optimal tool
by estimating the uncertainty of an MLLM's output. Our experiment shows that
the average improvement of SRICE over the base MLLM is 4.6% on five datasets
and the performance on some datasets even outperforms fine-tuning-based
methods, revealing the significance of ensuring reliable tool use in an MLLM
agent.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 11:18:53 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Zhi",
"Zhuo",
""
],
[
"Feng",
"Chen",
""
],
[
"Daneshmend",
"Adam",
""
],
[
"Orlu",
"Mine",
""
],
[
"Demosthenous",
"Andreas",
""
],
[
"Yin",
"Lu",
""
],
[
"Li",
"Da",
""
],
[
"Liu",
"Ziquan",
""
],
[
"Rodrigues",
"Miguel R. D.",
""
]
]
| TITLE: Seeing and Reasoning with Confidence: Supercharging Multimodal LLMs with
an Uncertainty-Aware Agentic Framework
ABSTRACT: Multimodal large language models (MLLMs) show promise in tasks like visual
question answering (VQA) but still face challenges in multimodal reasoning.
Recent works adapt agentic frameworks or chain-of-thought (CoT) reasoning to
improve performance. However, CoT-based multimodal reasoning often demands
costly data annotation and fine-tuning, while agentic approaches relying on
external tools risk introducing unreliable output from these tools. In this
paper, we propose Seeing and Reasoning with Confidence (SRICE), a training-free
multimodal reasoning framework that integrates external vision models with
uncertainty quantification (UQ) into an MLLM to address these challenges.
Specifically, SRICE guides the inference process by allowing MLLM to
autonomously select regions of interest through multi-stage interactions with
the help of external tools. We propose to use a conformal prediction-based
approach to calibrate the output of external tools and select the optimal tool
by estimating the uncertainty of an MLLM's output. Our experiment shows that
the average improvement of SRICE over the base MLLM is 4.6% on five datasets
and the performance on some datasets even outperforms fine-tuning-based
methods, revealing the significance of ensuring reliable tool use in an MLLM
agent.
| no_new_dataset | 0.946646 |
2503.08316 | Georgios Katranis | Georgios Katranis, Frederik Plahl, Joachim Grimstadt, Ilshat Mamaev,
Silvia Vock, Andrey Morozov | Dynamic Risk Assessment for Human-Robot Collaboration Using a
Heuristics-based Approach | null | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Human-robot collaboration (HRC) introduces significant safety challenges,
particularly in protecting human operators working alongside collaborative
robots (cobots). While current ISO standards emphasize risk assessment and
hazard identification, these procedures are often insufficient for addressing
the complexity of HRC environments, which involve numerous design factors and
dynamic interactions. This publication presents a method for objective hazard
analysis to support Dynamic Risk Assessment, extending beyond reliance on
expert knowledge. The approach monitors scene parameters, such as the distance
between human body parts and the cobot, as well as the cobot`s Cartesian
velocity. Additionally, an anthropocentric parameter focusing on the
orientation of the human head within the collaborative workspace is introduced.
These parameters are transformed into hazard indicators using non-linear
heuristic functions. The hazard indicators are then aggregated to estimate the
total hazard level of a given scenario. The proposed method is evaluated using
an industrial dataset that depicts various interactions between a human
operator and a cobot.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 11:25:47 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Katranis",
"Georgios",
""
],
[
"Plahl",
"Frederik",
""
],
[
"Grimstadt",
"Joachim",
""
],
[
"Mamaev",
"Ilshat",
""
],
[
"Vock",
"Silvia",
""
],
[
"Morozov",
"Andrey",
""
]
]
| TITLE: Dynamic Risk Assessment for Human-Robot Collaboration Using a
Heuristics-based Approach
ABSTRACT: Human-robot collaboration (HRC) introduces significant safety challenges,
particularly in protecting human operators working alongside collaborative
robots (cobots). While current ISO standards emphasize risk assessment and
hazard identification, these procedures are often insufficient for addressing
the complexity of HRC environments, which involve numerous design factors and
dynamic interactions. This publication presents a method for objective hazard
analysis to support Dynamic Risk Assessment, extending beyond reliance on
expert knowledge. The approach monitors scene parameters, such as the distance
between human body parts and the cobot, as well as the cobot`s Cartesian
velocity. Additionally, an anthropocentric parameter focusing on the
orientation of the human head within the collaborative workspace is introduced.
These parameters are transformed into hazard indicators using non-linear
heuristic functions. The hazard indicators are then aggregated to estimate the
total hazard level of a given scenario. The proposed method is evaluated using
an industrial dataset that depicts various interactions between a human
operator and a cobot.
| new_dataset | 0.963609 |
2503.08323 | Morteza Rohanian | Morteza Rohanian, Tarun Mehra, Nicola Miglino, Farhad Nooralahzadeh,
Michael Krauthammer, Andreas Wicki | Towards Scalable and Cross-Lingual Specialist Language Models for
Oncology | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Clinical oncology generates vast, unstructured data that often contain
inconsistencies, missing information, and ambiguities, making it difficult to
extract reliable insights for data-driven decision-making. General-purpose
large language models (LLMs) struggle with these challenges due to their lack
of domain-specific reasoning, including specialized clinical terminology,
context-dependent interpretations, and multi-modal data integration. We address
these issues with an oncology-specialized, efficient, and adaptable NLP
framework that combines instruction tuning, retrieval-augmented generation
(RAG), and graph-based knowledge integration. Our lightweight models prove
effective at oncology-specific tasks, such as named entity recognition (e.g.,
identifying cancer diagnoses), entity linking (e.g., linking entities to
standardized ontologies), TNM staging, document classification (e.g., cancer
subtype classification from pathology reports), and treatment response
prediction. Our framework emphasizes adaptability and resource efficiency. We
include minimal German instructions, collected at the University Hospital
Zurich (USZ), to test whether small amounts of non-English language data can
effectively transfer knowledge across languages. This approach mirrors our
motivation for lightweight models, which balance strong performance with
reduced computational costs, making them suitable for resource-limited
healthcare settings. We validated our models on oncology datasets,
demonstrating strong results in named entity recognition, relation extraction,
and document classification.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 11:34:57 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Rohanian",
"Morteza",
""
],
[
"Mehra",
"Tarun",
""
],
[
"Miglino",
"Nicola",
""
],
[
"Nooralahzadeh",
"Farhad",
""
],
[
"Krauthammer",
"Michael",
""
],
[
"Wicki",
"Andreas",
""
]
]
| TITLE: Towards Scalable and Cross-Lingual Specialist Language Models for
Oncology
ABSTRACT: Clinical oncology generates vast, unstructured data that often contain
inconsistencies, missing information, and ambiguities, making it difficult to
extract reliable insights for data-driven decision-making. General-purpose
large language models (LLMs) struggle with these challenges due to their lack
of domain-specific reasoning, including specialized clinical terminology,
context-dependent interpretations, and multi-modal data integration. We address
these issues with an oncology-specialized, efficient, and adaptable NLP
framework that combines instruction tuning, retrieval-augmented generation
(RAG), and graph-based knowledge integration. Our lightweight models prove
effective at oncology-specific tasks, such as named entity recognition (e.g.,
identifying cancer diagnoses), entity linking (e.g., linking entities to
standardized ontologies), TNM staging, document classification (e.g., cancer
subtype classification from pathology reports), and treatment response
prediction. Our framework emphasizes adaptability and resource efficiency. We
include minimal German instructions, collected at the University Hospital
Zurich (USZ), to test whether small amounts of non-English language data can
effectively transfer knowledge across languages. This approach mirrors our
motivation for lightweight models, which balance strong performance with
reduced computational costs, making them suitable for resource-limited
healthcare settings. We validated our models on oncology datasets,
demonstrating strong results in named entity recognition, relation extraction,
and document classification.
| no_new_dataset | 0.951504 |
2503.08328 | Liang Yu | Liang Yu, Lai Tu, Xiang Bai | MFRS: A Multi-Frequency Reference Series Approach to Scalable and
Accurate Time-Series Forecasting | null | null | null | null | cs.LG cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multivariate time-series forecasting holds immense value across diverse
applications, requiring methods to effectively capture complex temporal and
inter-variable dynamics. A key challenge lies in uncovering the intrinsic
patterns that govern predictability, beyond conventional designs, focusing on
network architectures to explore latent relationships or temporal dependencies.
Inspired by signal decomposition, this paper posits that time series
predictability is derived from periodic characteristics at different
frequencies. Consequently, we propose a novel time series forecasting method
based on multi-frequency reference series correlation analysis. Through
spectral analysis on long-term training data, we identify dominant spectral
components and their harmonics to design base-pattern reference series. Unlike
signal decomposition, which represents the original series as a linear
combination of basis signals, our method uses a transformer model to compute
cross-attention between the original series and reference series, capturing
essential features for forecasting. Experiments on major open and synthetic
datasets show state-of-the-art performance. Furthermore, by focusing on
attention with a small number of reference series rather than pairwise variable
attention, our method ensures scalability and broad applicability. The source
code is available at: https://github.com/yuliang555/MFRS
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 11:40:14 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Yu",
"Liang",
""
],
[
"Tu",
"Lai",
""
],
[
"Bai",
"Xiang",
""
]
]
| TITLE: MFRS: A Multi-Frequency Reference Series Approach to Scalable and
Accurate Time-Series Forecasting
ABSTRACT: Multivariate time-series forecasting holds immense value across diverse
applications, requiring methods to effectively capture complex temporal and
inter-variable dynamics. A key challenge lies in uncovering the intrinsic
patterns that govern predictability, beyond conventional designs, focusing on
network architectures to explore latent relationships or temporal dependencies.
Inspired by signal decomposition, this paper posits that time series
predictability is derived from periodic characteristics at different
frequencies. Consequently, we propose a novel time series forecasting method
based on multi-frequency reference series correlation analysis. Through
spectral analysis on long-term training data, we identify dominant spectral
components and their harmonics to design base-pattern reference series. Unlike
signal decomposition, which represents the original series as a linear
combination of basis signals, our method uses a transformer model to compute
cross-attention between the original series and reference series, capturing
essential features for forecasting. Experiments on major open and synthetic
datasets show state-of-the-art performance. Furthermore, by focusing on
attention with a small number of reference series rather than pairwise variable
attention, our method ensures scalability and broad applicability. The source
code is available at: https://github.com/yuliang555/MFRS
| no_new_dataset | 0.944022 |
2503.08335 | Soumya Jahagirdar | Soumya Shamarao Jahagirdar, Jayasree Saha, C V Jawahar | Prompt2LVideos: Exploring Prompts for Understanding Long-Form Multimodal
Videos | CVIP 2024 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Learning multimodal video understanding typically relies on datasets
comprising video clips paired with manually annotated captions. However, this
becomes even more challenging when dealing with long-form videos, lasting from
minutes to hours, in educational and news domains due to the need for more
annotators with subject expertise. Hence, there arises a need for automated
solutions. Recent advancements in Large Language Models (LLMs) promise to
capture concise and informative content that allows the comprehension of entire
videos by leveraging Automatic Speech Recognition (ASR) and Optical Character
Recognition (OCR) technologies. ASR provides textual content from audio, while
OCR extracts textual content from specific frames. This paper introduces a
dataset comprising long-form lectures and news videos. We present baseline
approaches to understand their limitations on this dataset and advocate for
exploring prompt engineering techniques to comprehend long-form multimodal
video datasets comprehensively.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 11:47:48 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Jahagirdar",
"Soumya Shamarao",
""
],
[
"Saha",
"Jayasree",
""
],
[
"Jawahar",
"C V",
""
]
]
| TITLE: Prompt2LVideos: Exploring Prompts for Understanding Long-Form Multimodal
Videos
ABSTRACT: Learning multimodal video understanding typically relies on datasets
comprising video clips paired with manually annotated captions. However, this
becomes even more challenging when dealing with long-form videos, lasting from
minutes to hours, in educational and news domains due to the need for more
annotators with subject expertise. Hence, there arises a need for automated
solutions. Recent advancements in Large Language Models (LLMs) promise to
capture concise and informative content that allows the comprehension of entire
videos by leveraging Automatic Speech Recognition (ASR) and Optical Character
Recognition (OCR) technologies. ASR provides textual content from audio, while
OCR extracts textual content from specific frames. This paper introduces a
dataset comprising long-form lectures and news videos. We present baseline
approaches to understand their limitations on this dataset and advocate for
exploring prompt engineering techniques to comprehend long-form multimodal
video datasets comprehensively.
| new_dataset | 0.96862 |
2503.08336 | Runwei Guan | Runwei Guan, Jianan Liu, Ningwei Ouyang, Daizong Liu, Xiaolou Sun,
Lianqing Zheng, Ming Xu, Yutao Yue, Hui Xiong | Talk2PC: Enhancing 3D Visual Grounding through LiDAR and Radar Point
Clouds Fusion for Autonomous Driving | 14 pages, 11 figures | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Embodied outdoor scene understanding forms the foundation for autonomous
agents to perceive, analyze, and react to dynamic driving environments.
However, existing 3D understanding is predominantly based on 2D Vision-Language
Models (VLMs), collecting and processing limited scene-aware contexts. Instead,
compared to the 2D planar visual information, point cloud sensors like LiDAR
offer rich depth information and fine-grained 3D representations of objects.
Meanwhile, the emerging 4D millimeter-wave (mmWave) radar is capable of
detecting the motion trend, velocity, and reflection intensity of each object.
Therefore, the integration of these two modalities provides more flexible
querying conditions for natural language, enabling more accurate 3D visual
grounding. To this end, in this paper, we exploratively propose a novel method
called TPCNet, the first outdoor 3D visual grounding model upon the paradigm of
prompt-guided point cloud sensor combination, including both LiDAR and radar
contexts. To adaptively balance the features of these two sensors required by
the prompt, we have designed a multi-fusion paradigm called Two-Stage
Heterogeneous Modal Adaptive Fusion. Specifically, this paradigm initially
employs Bidirectional Agent Cross-Attention (BACA), which feeds dual-sensor
features, characterized by global receptive fields, to the text features for
querying. Additionally, we have designed a Dynamic Gated Graph Fusion (DGGF)
module to locate the regions of interest identified by the queries. To further
enhance accuracy, we innovatively devise an C3D-RECHead, based on the nearest
object edge. Our experiments have demonstrated that our TPCNet, along with its
individual modules, achieves the state-of-the-art performance on both the
Talk2Radar and Talk2Car datasets.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 11:48:27 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Guan",
"Runwei",
""
],
[
"Liu",
"Jianan",
""
],
[
"Ouyang",
"Ningwei",
""
],
[
"Liu",
"Daizong",
""
],
[
"Sun",
"Xiaolou",
""
],
[
"Zheng",
"Lianqing",
""
],
[
"Xu",
"Ming",
""
],
[
"Yue",
"Yutao",
""
],
[
"Xiong",
"Hui",
""
]
]
| TITLE: Talk2PC: Enhancing 3D Visual Grounding through LiDAR and Radar Point
Clouds Fusion for Autonomous Driving
ABSTRACT: Embodied outdoor scene understanding forms the foundation for autonomous
agents to perceive, analyze, and react to dynamic driving environments.
However, existing 3D understanding is predominantly based on 2D Vision-Language
Models (VLMs), collecting and processing limited scene-aware contexts. Instead,
compared to the 2D planar visual information, point cloud sensors like LiDAR
offer rich depth information and fine-grained 3D representations of objects.
Meanwhile, the emerging 4D millimeter-wave (mmWave) radar is capable of
detecting the motion trend, velocity, and reflection intensity of each object.
Therefore, the integration of these two modalities provides more flexible
querying conditions for natural language, enabling more accurate 3D visual
grounding. To this end, in this paper, we exploratively propose a novel method
called TPCNet, the first outdoor 3D visual grounding model upon the paradigm of
prompt-guided point cloud sensor combination, including both LiDAR and radar
contexts. To adaptively balance the features of these two sensors required by
the prompt, we have designed a multi-fusion paradigm called Two-Stage
Heterogeneous Modal Adaptive Fusion. Specifically, this paradigm initially
employs Bidirectional Agent Cross-Attention (BACA), which feeds dual-sensor
features, characterized by global receptive fields, to the text features for
querying. Additionally, we have designed a Dynamic Gated Graph Fusion (DGGF)
module to locate the regions of interest identified by the queries. To further
enhance accuracy, we innovatively devise an C3D-RECHead, based on the nearest
object edge. Our experiments have demonstrated that our TPCNet, along with its
individual modules, achieves the state-of-the-art performance on both the
Talk2Radar and Talk2Car datasets.
| no_new_dataset | 0.948058 |
2503.08346 | Chanyoung Kim | Chanyoung Kim, Dayun Ju, Jinyeong Kim, Woojung Han, Roberto
Alcover-Couso and Seong Jae Hwang | Pathology-Aware Adaptive Watermarking for Text-Driven Medical Image
Synthesis | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | As recent text-conditioned diffusion models have enabled the generation of
high-quality images, concerns over their potential misuse have also grown. This
issue is critical in the medical domain, where text-conditioned generated
medical images could enable insurance fraud or falsified records, highlighting
the urgent need for reliable safeguards against unethical use. While
watermarking techniques have emerged as a promising solution in general image
domains, their direct application to medical imaging presents significant
challenges. A key challenge is preserving fine-grained disease manifestations,
as even minor distortions from a watermark may lead to clinical
misinterpretation, which compromises diagnostic integrity. To overcome this
gap, we present MedSign, a deep learning-based watermarking framework
specifically designed for text-to-medical image synthesis, which preserves
pathologically significant regions by adaptively adjusting watermark strength.
Specifically, we generate a pathology localization map using cross-attention
between medical text tokens and the diffusion denoising network, aggregating
token-wise attention across layers, heads, and time steps. Leveraging this map,
we optimize the LDM decoder to incorporate watermarking during image synthesis,
ensuring cohesive integration while minimizing interference in diagnostically
critical regions. Experimental results show that our MedSign preserves
diagnostic integrity while ensuring watermark robustness, achieving
state-of-the-art performance in image quality and detection accuracy on
MIMIC-CXR and OIA-ODIR datasets.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 11:55:14 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Kim",
"Chanyoung",
""
],
[
"Ju",
"Dayun",
""
],
[
"Kim",
"Jinyeong",
""
],
[
"Han",
"Woojung",
""
],
[
"Alcover-Couso",
"Roberto",
""
],
[
"Hwang",
"Seong Jae",
""
]
]
| TITLE: Pathology-Aware Adaptive Watermarking for Text-Driven Medical Image
Synthesis
ABSTRACT: As recent text-conditioned diffusion models have enabled the generation of
high-quality images, concerns over their potential misuse have also grown. This
issue is critical in the medical domain, where text-conditioned generated
medical images could enable insurance fraud or falsified records, highlighting
the urgent need for reliable safeguards against unethical use. While
watermarking techniques have emerged as a promising solution in general image
domains, their direct application to medical imaging presents significant
challenges. A key challenge is preserving fine-grained disease manifestations,
as even minor distortions from a watermark may lead to clinical
misinterpretation, which compromises diagnostic integrity. To overcome this
gap, we present MedSign, a deep learning-based watermarking framework
specifically designed for text-to-medical image synthesis, which preserves
pathologically significant regions by adaptively adjusting watermark strength.
Specifically, we generate a pathology localization map using cross-attention
between medical text tokens and the diffusion denoising network, aggregating
token-wise attention across layers, heads, and time steps. Leveraging this map,
we optimize the LDM decoder to incorporate watermarking during image synthesis,
ensuring cohesive integration while minimizing interference in diagnostically
critical regions. Experimental results show that our MedSign preserves
diagnostic integrity while ensuring watermark robustness, achieving
state-of-the-art performance in image quality and detection accuracy on
MIMIC-CXR and OIA-ODIR datasets.
| no_new_dataset | 0.943452 |
2503.08348 | Sangram Patil | H. P. Khandagale, Sangram Patil, V. S. Gavali, S. V. Chavan, P. P.
Halkarnikar, Prateek A. Meshram | Design and Implementation of FourCropNet: A CNN-Based System for
Efficient Multi-Crop Disease Detection and Management | null | Journal of Information Systems Engineering and Management 2025,
10(7s) e-ISSN: 2468-4376 | 10.52783/jisem.v10i7s.877 | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Plant disease detection is a critical task in agriculture, directly impacting
crop yield, food security, and sustainable farming practices. This study
proposes FourCropNet, a novel deep learning model designed to detect diseases
in multiple crops, including CottonLeaf, Grape, Soybean, and Corn. The model
leverages an advanced architecture comprising residual blocks for efficient
feature extraction, attention mechanisms to enhance focus on disease-relevant
regions, and lightweight layers for computational efficiency. These components
collectively enable FourCropNet to achieve superior performance across varying
datasets and class complexities, from single-crop datasets to combined datasets
with 15 classes. The proposed model was evaluated on diverse datasets,
demonstrating high accuracy, specificity, sensitivity, and F1 scores. Notably,
FourCropNet achieved the highest accuracy of 99.7% for Grape, 99.5% for Corn,
and 95.3% for the combined dataset. Its scalability and ability to generalize
across datasets underscore its robustness. Comparative analysis shows that
FourCropNet consistently outperforms state-of-the-art models such as MobileNet,
VGG16, and EfficientNet across various metrics. FourCropNet's innovative design
and consistent performance make it a reliable solution for real-time disease
detection in agriculture. This model has the potential to assist farmers in
timely disease diagnosis, reducing economic losses and promoting sustainable
agricultural practices.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 12:00:56 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Khandagale",
"H. P.",
""
],
[
"Patil",
"Sangram",
""
],
[
"Gavali",
"V. S.",
""
],
[
"Chavan",
"S. V.",
""
],
[
"Halkarnikar",
"P. P.",
""
],
[
"Meshram",
"Prateek A.",
""
]
]
| TITLE: Design and Implementation of FourCropNet: A CNN-Based System for
Efficient Multi-Crop Disease Detection and Management
ABSTRACT: Plant disease detection is a critical task in agriculture, directly impacting
crop yield, food security, and sustainable farming practices. This study
proposes FourCropNet, a novel deep learning model designed to detect diseases
in multiple crops, including CottonLeaf, Grape, Soybean, and Corn. The model
leverages an advanced architecture comprising residual blocks for efficient
feature extraction, attention mechanisms to enhance focus on disease-relevant
regions, and lightweight layers for computational efficiency. These components
collectively enable FourCropNet to achieve superior performance across varying
datasets and class complexities, from single-crop datasets to combined datasets
with 15 classes. The proposed model was evaluated on diverse datasets,
demonstrating high accuracy, specificity, sensitivity, and F1 scores. Notably,
FourCropNet achieved the highest accuracy of 99.7% for Grape, 99.5% for Corn,
and 95.3% for the combined dataset. Its scalability and ability to generalize
across datasets underscore its robustness. Comparative analysis shows that
FourCropNet consistently outperforms state-of-the-art models such as MobileNet,
VGG16, and EfficientNet across various metrics. FourCropNet's innovative design
and consistent performance make it a reliable solution for real-time disease
detection in agriculture. This model has the potential to assist farmers in
timely disease diagnosis, reducing economic losses and promoting sustainable
agricultural practices.
| no_new_dataset | 0.945197 |
2503.08358 | Md Faizal Karim | Md Faizal Karim, Mohammed Saad Hashmi, Shreya Bollimuntha, Mahesh
Reddy Tapeti, Gaurav Singh, Nagamanikandan Govindan, K Madhava Krishna | DG16M: A Large-Scale Dataset for Dual-Arm Grasping with Force-Optimized
Grasps | null | null | null | null | cs.RO | http://creativecommons.org/licenses/by/4.0/ | Dual-arm robotic grasping is crucial for handling large objects that require
stable and coordinated manipulation. While single-arm grasping has been
extensively studied, datasets tailored for dual-arm settings remain scarce. We
introduce a large-scale dataset of 16 million dual-arm grasps, evaluated under
improved force-closure constraints. Additionally, we develop a benchmark
dataset containing 300 objects with approximately 30,000 grasps, evaluated in a
physics simulation environment, providing a better grasp quality assessment for
dual-arm grasp synthesis methods. Finally, we demonstrate the effectiveness of
our dataset by training a Dual-Arm Grasp Classifier network that outperforms
the state-of-the-art methods by 15\%, achieving higher grasp success rates and
improved generalization across objects.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 12:15:20 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Karim",
"Md Faizal",
""
],
[
"Hashmi",
"Mohammed Saad",
""
],
[
"Bollimuntha",
"Shreya",
""
],
[
"Tapeti",
"Mahesh Reddy",
""
],
[
"Singh",
"Gaurav",
""
],
[
"Govindan",
"Nagamanikandan",
""
],
[
"Krishna",
"K Madhava",
""
]
]
| TITLE: DG16M: A Large-Scale Dataset for Dual-Arm Grasping with Force-Optimized
Grasps
ABSTRACT: Dual-arm robotic grasping is crucial for handling large objects that require
stable and coordinated manipulation. While single-arm grasping has been
extensively studied, datasets tailored for dual-arm settings remain scarce. We
introduce a large-scale dataset of 16 million dual-arm grasps, evaluated under
improved force-closure constraints. Additionally, we develop a benchmark
dataset containing 300 objects with approximately 30,000 grasps, evaluated in a
physics simulation environment, providing a better grasp quality assessment for
dual-arm grasp synthesis methods. Finally, we demonstrate the effectiveness of
our dataset by training a Dual-Arm Grasp Classifier network that outperforms
the state-of-the-art methods by 15\%, achieving higher grasp success rates and
improved generalization across objects.
| new_dataset | 0.958731 |
2503.08363 | Zhaiyu Chen | Zhaiyu Chen, Yuqing Wang, Liangliang Nan, Xiao Xiang Zhu | Parametric Point Cloud Completion for Polygonal Surface Reconstruction | CVPR 2025 | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Existing polygonal surface reconstruction methods heavily depend on input
completeness and struggle with incomplete point clouds. We argue that while
current point cloud completion techniques may recover missing points, they are
not optimized for polygonal surface reconstruction, where the parametric
representation of underlying surfaces remains overlooked. To address this gap,
we introduce parametric completion, a novel paradigm for point cloud
completion, which recovers parametric primitives instead of individual points
to convey high-level geometric structures. Our presented approach, PaCo,
enables high-quality polygonal surface reconstruction by leveraging plane
proxies that encapsulate both plane parameters and inlier points, proving
particularly effective in challenging scenarios with highly incomplete data.
Comprehensive evaluations of our approach on the ABC dataset establish its
effectiveness with superior performance and set a new standard for polygonal
surface reconstruction from incomplete data. Project page:
https://parametric-completion.github.io.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 12:20:24 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Chen",
"Zhaiyu",
""
],
[
"Wang",
"Yuqing",
""
],
[
"Nan",
"Liangliang",
""
],
[
"Zhu",
"Xiao Xiang",
""
]
]
| TITLE: Parametric Point Cloud Completion for Polygonal Surface Reconstruction
ABSTRACT: Existing polygonal surface reconstruction methods heavily depend on input
completeness and struggle with incomplete point clouds. We argue that while
current point cloud completion techniques may recover missing points, they are
not optimized for polygonal surface reconstruction, where the parametric
representation of underlying surfaces remains overlooked. To address this gap,
we introduce parametric completion, a novel paradigm for point cloud
completion, which recovers parametric primitives instead of individual points
to convey high-level geometric structures. Our presented approach, PaCo,
enables high-quality polygonal surface reconstruction by leveraging plane
proxies that encapsulate both plane parameters and inlier points, proving
particularly effective in challenging scenarios with highly incomplete data.
Comprehensive evaluations of our approach on the ABC dataset establish its
effectiveness with superior performance and set a new standard for polygonal
surface reconstruction from incomplete data. Project page:
https://parametric-completion.github.io.
| no_new_dataset | 0.952618 |
2503.08367 | Runling Long | Runling Long, Yunlong Wang, Jia Wan, Xiang Deng, Xinting Zhu, Weili
Guan, Antoni B. Chan, Liqiang Nie | Embodied Crowd Counting | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Occlusion is one of the fundamental challenges in crowd counting. In the
community, various data-driven approaches have been developed to address this
issue, yet their effectiveness is limited. This is mainly because most existing
crowd counting datasets on which the methods are trained are based on passive
cameras, restricting their ability to fully sense the environment. Recently,
embodied navigation methods have shown significant potential in precise object
detection in interactive scenes. These methods incorporate active camera
settings, holding promise in addressing the fundamental issues in crowd
counting. However, most existing methods are designed for indoor navigation,
showing unknown performance in analyzing complex object distribution in large
scale scenes, such as crowds. Besides, most existing embodied navigation
datasets are indoor scenes with limited scale and object quantity, preventing
them from being introduced into dense crowd analysis. Based on this, a novel
task, Embodied Crowd Counting (ECC), is proposed. We first build up an
interactive simulator, Embodied Crowd Counting Dataset (ECCD), which enables
large scale scenes and large object quantity. A prior probability distribution
that approximates realistic crowd distribution is introduced to generate
crowds. Then, a zero-shot navigation method (ZECC) is proposed. This method
contains a MLLM driven coarse-to-fine navigation mechanism, enabling active
Z-axis exploration, and a normal-line-based crowd distribution analysis method
for fine counting. Experimental results against baselines show that the
proposed method achieves the best trade-off between counting accuracy and
navigation cost.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 12:23:34 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Long",
"Runling",
""
],
[
"Wang",
"Yunlong",
""
],
[
"Wan",
"Jia",
""
],
[
"Deng",
"Xiang",
""
],
[
"Zhu",
"Xinting",
""
],
[
"Guan",
"Weili",
""
],
[
"Chan",
"Antoni B.",
""
],
[
"Nie",
"Liqiang",
""
]
]
| TITLE: Embodied Crowd Counting
ABSTRACT: Occlusion is one of the fundamental challenges in crowd counting. In the
community, various data-driven approaches have been developed to address this
issue, yet their effectiveness is limited. This is mainly because most existing
crowd counting datasets on which the methods are trained are based on passive
cameras, restricting their ability to fully sense the environment. Recently,
embodied navigation methods have shown significant potential in precise object
detection in interactive scenes. These methods incorporate active camera
settings, holding promise in addressing the fundamental issues in crowd
counting. However, most existing methods are designed for indoor navigation,
showing unknown performance in analyzing complex object distribution in large
scale scenes, such as crowds. Besides, most existing embodied navigation
datasets are indoor scenes with limited scale and object quantity, preventing
them from being introduced into dense crowd analysis. Based on this, a novel
task, Embodied Crowd Counting (ECC), is proposed. We first build up an
interactive simulator, Embodied Crowd Counting Dataset (ECCD), which enables
large scale scenes and large object quantity. A prior probability distribution
that approximates realistic crowd distribution is introduced to generate
crowds. Then, a zero-shot navigation method (ZECC) is proposed. This method
contains a MLLM driven coarse-to-fine navigation mechanism, enabling active
Z-axis exploration, and a normal-line-based crowd distribution analysis method
for fine counting. Experimental results against baselines show that the
proposed method achieves the best trade-off between counting accuracy and
navigation cost.
| new_dataset | 0.841956 |
2503.08368 | Chaoquan Jiang | Chaoquan Jiang, Yunfan Yang, Rui Hu, Jitao Sang | Debiased Prompt Tuning in Vision-Language Model without Annotations | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Prompt tuning of Vision-Language Models (VLMs) such as CLIP, has demonstrated
the ability to rapidly adapt to various downstream tasks. However, recent
studies indicate that tuned VLMs may suffer from the problem of spurious
correlations, where the model relies on spurious features (e.g. background and
gender) in the data. This may lead to the model having worse robustness in
out-of-distribution data. Standard methods for eliminating spurious correlation
typically require us to know the spurious attribute labels of each sample,
which is hard in the real world. In this work, we explore improving the group
robustness of prompt tuning in VLMs without relying on manual annotation of
spurious features. We notice the zero - shot image recognition ability of VLMs
and use this ability to identify spurious features, thus avoiding the cost of
manual annotation. By leveraging pseudo-spurious attribute annotations, we
further propose a method to automatically adjust the training weights of
different groups. Extensive experiments show that our approach efficiently
improves the worst-group accuracy on CelebA, Waterbirds, and MetaShift
datasets, achieving the best robustness gap between the worst-group accuracy
and the overall accuracy.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 12:24:54 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Jiang",
"Chaoquan",
""
],
[
"Yang",
"Yunfan",
""
],
[
"Hu",
"Rui",
""
],
[
"Sang",
"Jitao",
""
]
]
| TITLE: Debiased Prompt Tuning in Vision-Language Model without Annotations
ABSTRACT: Prompt tuning of Vision-Language Models (VLMs) such as CLIP, has demonstrated
the ability to rapidly adapt to various downstream tasks. However, recent
studies indicate that tuned VLMs may suffer from the problem of spurious
correlations, where the model relies on spurious features (e.g. background and
gender) in the data. This may lead to the model having worse robustness in
out-of-distribution data. Standard methods for eliminating spurious correlation
typically require us to know the spurious attribute labels of each sample,
which is hard in the real world. In this work, we explore improving the group
robustness of prompt tuning in VLMs without relying on manual annotation of
spurious features. We notice the zero - shot image recognition ability of VLMs
and use this ability to identify spurious features, thus avoiding the cost of
manual annotation. By leveraging pseudo-spurious attribute annotations, we
further propose a method to automatically adjust the training weights of
different groups. Extensive experiments show that our approach efficiently
improves the worst-group accuracy on CelebA, Waterbirds, and MetaShift
datasets, achieving the best robustness gap between the worst-group accuracy
and the overall accuracy.
| no_new_dataset | 0.948489 |
2503.08370 | Xucheng Guo | Xucheng Guo, Yiran Shen, Xiaofang Xiao, Yuanfeng Zhou, Lin Wang | Ev-Layout: A Large-scale Event-based Multi-modal Dataset for Indoor
Layout Estimation and Tracking | null | null | null | null | cs.GR cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | This paper presents Ev-Layout, a novel large-scale event-based multi-modal
dataset designed for indoor layout estimation and tracking. Ev-Layout makes key
contributions to the community by: Utilizing a hybrid data collection platform
(with a head-mounted display and VR interface) that integrates both RGB and
bio-inspired event cameras to capture indoor layouts in motion. Incorporating
time-series data from inertial measurement units (IMUs) and ambient lighting
conditions recorded during data collection to highlight the potential impact of
motion speed and lighting on layout estimation accuracy. The dataset consists
of 2.5K sequences, including over 771.3K RGB images and 10 billion event data
points. Of these, 39K images are annotated with indoor layouts, enabling
research in both event-based and video-based indoor layout estimation. Based on
the dataset, we propose an event-based layout estimation pipeline with a novel
event-temporal distribution feature module to effectively aggregate the
spatio-temporal information from events. Additionally, we introduce a
spatio-temporal feature fusion module that can be easily integrated into a
transformer module for fusion purposes. Finally, we conduct benchmarking and
extensive experiments on the Ev-Layout dataset, demonstrating that our approach
significantly improves the accuracy of dynamic indoor layout estimation
compared to existing event-based methods.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 12:26:39 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Guo",
"Xucheng",
""
],
[
"Shen",
"Yiran",
""
],
[
"Xiao",
"Xiaofang",
""
],
[
"Zhou",
"Yuanfeng",
""
],
[
"Wang",
"Lin",
""
]
]
| TITLE: Ev-Layout: A Large-scale Event-based Multi-modal Dataset for Indoor
Layout Estimation and Tracking
ABSTRACT: This paper presents Ev-Layout, a novel large-scale event-based multi-modal
dataset designed for indoor layout estimation and tracking. Ev-Layout makes key
contributions to the community by: Utilizing a hybrid data collection platform
(with a head-mounted display and VR interface) that integrates both RGB and
bio-inspired event cameras to capture indoor layouts in motion. Incorporating
time-series data from inertial measurement units (IMUs) and ambient lighting
conditions recorded during data collection to highlight the potential impact of
motion speed and lighting on layout estimation accuracy. The dataset consists
of 2.5K sequences, including over 771.3K RGB images and 10 billion event data
points. Of these, 39K images are annotated with indoor layouts, enabling
research in both event-based and video-based indoor layout estimation. Based on
the dataset, we propose an event-based layout estimation pipeline with a novel
event-temporal distribution feature module to effectively aggregate the
spatio-temporal information from events. Additionally, we introduce a
spatio-temporal feature fusion module that can be easily integrated into a
transformer module for fusion purposes. Finally, we conduct benchmarking and
extensive experiments on the Ev-Layout dataset, demonstrating that our approach
significantly improves the accuracy of dynamic indoor layout estimation
compared to existing event-based methods.
| new_dataset | 0.968171 |
2503.08371 | Bariscan Bozkurt | Bariscan Bozkurt, Ben Deaner, Dimitri Meunier, Liyuan Xu, Arthur
Gretton | Density Ratio-based Proxy Causal Learning Without Density Ratios | AISTATS 2025 accepted, 81 pages | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | We address the setting of Proxy Causal Learning (PCL), which has the goal of
estimating causal effects from observed data in the presence of hidden
confounding. Proxy methods accomplish this task using two proxy variables
related to the latent confounder: a treatment proxy (related to the treatment)
and an outcome proxy (related to the outcome). Two approaches have been
proposed to perform causal effect estimation given proxy variables; however
only one of these has found mainstream acceptance, since the other was
understood to require density ratio estimation - a challenging task in high
dimensions. In the present work, we propose a practical and effective
implementation of the second approach, which bypasses explicit density ratio
estimation and is suitable for continuous and high-dimensional treatments. We
employ kernel ridge regression to derive estimators, resulting in simple
closed-form solutions for dose-response and conditional dose-response curves,
along with consistency guarantees. Our methods empirically demonstrate superior
or comparable performance to existing frameworks on synthetic and real-world
datasets.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 12:27:54 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Bozkurt",
"Bariscan",
""
],
[
"Deaner",
"Ben",
""
],
[
"Meunier",
"Dimitri",
""
],
[
"Xu",
"Liyuan",
""
],
[
"Gretton",
"Arthur",
""
]
]
| TITLE: Density Ratio-based Proxy Causal Learning Without Density Ratios
ABSTRACT: We address the setting of Proxy Causal Learning (PCL), which has the goal of
estimating causal effects from observed data in the presence of hidden
confounding. Proxy methods accomplish this task using two proxy variables
related to the latent confounder: a treatment proxy (related to the treatment)
and an outcome proxy (related to the outcome). Two approaches have been
proposed to perform causal effect estimation given proxy variables; however
only one of these has found mainstream acceptance, since the other was
understood to require density ratio estimation - a challenging task in high
dimensions. In the present work, we propose a practical and effective
implementation of the second approach, which bypasses explicit density ratio
estimation and is suitable for continuous and high-dimensional treatments. We
employ kernel ridge regression to derive estimators, resulting in simple
closed-form solutions for dose-response and conditional dose-response curves,
along with consistency guarantees. Our methods empirically demonstrate superior
or comparable performance to existing frameworks on synthetic and real-world
datasets.
| no_new_dataset | 0.949342 |
2503.08373 | Fabian Isensee | Fabian Isensee, Maximilian Rokuss, Lars Kr\"amer, Stefan Dinkelacker,
Ashis Ravindran, Florian Stritzke, Benjamin Hamm, Tassilo Wald, Moritz
Langenberg, Constantin Ulrich, Jonathan Deissler, Ralf Floca, Klaus
Maier-Hein | nnInteractive: Redefining 3D Promptable Segmentation | Fabian Isensee, Maximilian Rokuss and Lars Kr\"amer contributed
equally. Each co-first author may list themselves as lead author on their CV | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Accurate and efficient 3D segmentation is essential for both clinical and
research applications. While foundation models like SAM have revolutionized
interactive segmentation, their 2D design and domain shift limitations make
them ill-suited for 3D medical images. Current adaptations address some of
these challenges but remain limited, either lacking volumetric awareness,
offering restricted interactivity, or supporting only a small set of structures
and modalities. Usability also remains a challenge, as current tools are rarely
integrated into established imaging platforms and often rely on cumbersome
web-based interfaces with restricted functionality. We introduce nnInteractive,
the first comprehensive 3D interactive open-set segmentation method. It
supports diverse prompts-including points, scribbles, boxes, and a novel lasso
prompt-while leveraging intuitive 2D interactions to generate full 3D
segmentations. Trained on 120+ diverse volumetric 3D datasets (CT, MRI, PET, 3D
Microscopy, etc.), nnInteractive sets a new state-of-the-art in accuracy,
adaptability, and usability. Crucially, it is the first method integrated into
widely used image viewers (e.g., Napari, MITK), ensuring broad accessibility
for real-world clinical and research applications. Extensive benchmarking
demonstrates that nnInteractive far surpasses existing methods, setting a new
standard for AI-driven interactive 3D segmentation. nnInteractive is publicly
available: https://github.com/MIC-DKFZ/napari-nninteractive (Napari plugin),
https://www.mitk.org/MITK-nnInteractive (MITK integration),
https://github.com/MIC-DKFZ/nnInteractive (Python backend).
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 12:30:34 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Isensee",
"Fabian",
""
],
[
"Rokuss",
"Maximilian",
""
],
[
"Krämer",
"Lars",
""
],
[
"Dinkelacker",
"Stefan",
""
],
[
"Ravindran",
"Ashis",
""
],
[
"Stritzke",
"Florian",
""
],
[
"Hamm",
"Benjamin",
""
],
[
"Wald",
"Tassilo",
""
],
[
"Langenberg",
"Moritz",
""
],
[
"Ulrich",
"Constantin",
""
],
[
"Deissler",
"Jonathan",
""
],
[
"Floca",
"Ralf",
""
],
[
"Maier-Hein",
"Klaus",
""
]
]
| TITLE: nnInteractive: Redefining 3D Promptable Segmentation
ABSTRACT: Accurate and efficient 3D segmentation is essential for both clinical and
research applications. While foundation models like SAM have revolutionized
interactive segmentation, their 2D design and domain shift limitations make
them ill-suited for 3D medical images. Current adaptations address some of
these challenges but remain limited, either lacking volumetric awareness,
offering restricted interactivity, or supporting only a small set of structures
and modalities. Usability also remains a challenge, as current tools are rarely
integrated into established imaging platforms and often rely on cumbersome
web-based interfaces with restricted functionality. We introduce nnInteractive,
the first comprehensive 3D interactive open-set segmentation method. It
supports diverse prompts-including points, scribbles, boxes, and a novel lasso
prompt-while leveraging intuitive 2D interactions to generate full 3D
segmentations. Trained on 120+ diverse volumetric 3D datasets (CT, MRI, PET, 3D
Microscopy, etc.), nnInteractive sets a new state-of-the-art in accuracy,
adaptability, and usability. Crucially, it is the first method integrated into
widely used image viewers (e.g., Napari, MITK), ensuring broad accessibility
for real-world clinical and research applications. Extensive benchmarking
demonstrates that nnInteractive far surpasses existing methods, setting a new
standard for AI-driven interactive 3D segmentation. nnInteractive is publicly
available: https://github.com/MIC-DKFZ/napari-nninteractive (Napari plugin),
https://www.mitk.org/MITK-nnInteractive (MITK integration),
https://github.com/MIC-DKFZ/nnInteractive (Python backend).
| no_new_dataset | 0.944485 |
2503.08379 | Leandro Car\'isio Fernandes | Leandro Car\'isio Fernandes, Leandro dos Santos Ribeiro, Marcos
Vin\'icius Borela de Castro, Leonardo Augusto da Silva Pacheco, Edans
Fl\'avius de Oliveira Sandes | JurisTCU: A Brazilian Portuguese Information Retrieval Dataset with
Query Relevance Judgments | 21 pages | null | null | null | cs.IR cs.CL | http://creativecommons.org/licenses/by/4.0/ | This paper introduces JurisTCU, a Brazilian Portuguese dataset for legal
information retrieval (LIR). The dataset is freely available and consists of
16,045 jurisprudential documents from the Brazilian Federal Court of Accounts,
along with 150 queries annotated with relevance judgments. It addresses the
scarcity of Portuguese-language LIR datasets with query relevance annotations.
The queries are organized into three groups: real user keyword-based queries,
synthetic keyword-based queries, and synthetic question-based queries.
Relevance judgments were produced through a hybrid approach combining LLM-based
scoring with expert domain validation. We used JurisTCU in 14 experiments using
lexical search (document expansion methods) and semantic search (BERT-based and
OpenAI embeddings). We show that the document expansion methods significantly
improve the performance of standard BM25 search on this dataset, with
improvements exceeding 45% in P@10, R@10, and nDCG@10 metrics when evaluating
short keyword-based queries. Among the embedding models, the OpenAI models
produced the best results, with improvements of approximately 70% in P@10,
R@10, and nDCG@10 metrics for short keyword-based queries, suggesting that
these dense embeddings capture semantic relationships in this domain,
surpassing the reliance on lexical terms. Besides offering a dataset for the
Portuguese-language IR research community, suitable for evaluating search
systems, the results also contribute to enhancing a search system highly
relevant to Brazilian citizens.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 12:39:04 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Fernandes",
"Leandro Carísio",
""
],
[
"Ribeiro",
"Leandro dos Santos",
""
],
[
"de Castro",
"Marcos Vinícius Borela",
""
],
[
"Pacheco",
"Leonardo Augusto da Silva",
""
],
[
"Sandes",
"Edans Flávius de Oliveira",
""
]
]
| TITLE: JurisTCU: A Brazilian Portuguese Information Retrieval Dataset with
Query Relevance Judgments
ABSTRACT: This paper introduces JurisTCU, a Brazilian Portuguese dataset for legal
information retrieval (LIR). The dataset is freely available and consists of
16,045 jurisprudential documents from the Brazilian Federal Court of Accounts,
along with 150 queries annotated with relevance judgments. It addresses the
scarcity of Portuguese-language LIR datasets with query relevance annotations.
The queries are organized into three groups: real user keyword-based queries,
synthetic keyword-based queries, and synthetic question-based queries.
Relevance judgments were produced through a hybrid approach combining LLM-based
scoring with expert domain validation. We used JurisTCU in 14 experiments using
lexical search (document expansion methods) and semantic search (BERT-based and
OpenAI embeddings). We show that the document expansion methods significantly
improve the performance of standard BM25 search on this dataset, with
improvements exceeding 45% in P@10, R@10, and nDCG@10 metrics when evaluating
short keyword-based queries. Among the embedding models, the OpenAI models
produced the best results, with improvements of approximately 70% in P@10,
R@10, and nDCG@10 metrics for short keyword-based queries, suggesting that
these dense embeddings capture semantic relationships in this domain,
surpassing the reliance on lexical terms. Besides offering a dataset for the
Portuguese-language IR research community, suitable for evaluating search
systems, the results also contribute to enhancing a search system highly
relevant to Brazilian citizens.
| new_dataset | 0.976423 |
2503.08382 | Jesus Zarzar | Jesus Zarzar, Tom Monnier, Roman Shapovalov, Andrea Vedaldi, David
Novotny | Twinner: Shining Light on Digital Twins in a Few Snaps | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | We present the first large reconstruction model, Twinner, capable of
recovering a scene's illumination as well as an object's geometry and material
properties from only a few posed images. Twinner is based on the Large
Reconstruction Model and innovates in three key ways: 1) We introduce a
memory-efficient voxel-grid transformer whose memory scales only quadratically
with the size of the voxel grid. 2) To deal with scarcity of high-quality
ground-truth PBR-shaded models, we introduce a large fully-synthetic dataset of
procedurally-generated PBR-textured objects lit with varied illumination. 3) To
narrow the synthetic-to-real gap, we finetune the model on real life datasets
by means of a differentiable physically-based shading model, eschewing the need
for ground-truth illumination or material properties which are challenging to
obtain in real life. We demonstrate the efficacy of our model on the real life
StanfordORB benchmark where, given few input views, we achieve reconstruction
quality significantly superior to existing feedforward reconstruction networks,
and comparable to significantly slower per-scene optimization methods.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 12:43:11 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Zarzar",
"Jesus",
""
],
[
"Monnier",
"Tom",
""
],
[
"Shapovalov",
"Roman",
""
],
[
"Vedaldi",
"Andrea",
""
],
[
"Novotny",
"David",
""
]
]
| TITLE: Twinner: Shining Light on Digital Twins in a Few Snaps
ABSTRACT: We present the first large reconstruction model, Twinner, capable of
recovering a scene's illumination as well as an object's geometry and material
properties from only a few posed images. Twinner is based on the Large
Reconstruction Model and innovates in three key ways: 1) We introduce a
memory-efficient voxel-grid transformer whose memory scales only quadratically
with the size of the voxel grid. 2) To deal with scarcity of high-quality
ground-truth PBR-shaded models, we introduce a large fully-synthetic dataset of
procedurally-generated PBR-textured objects lit with varied illumination. 3) To
narrow the synthetic-to-real gap, we finetune the model on real life datasets
by means of a differentiable physically-based shading model, eschewing the need
for ground-truth illumination or material properties which are challenging to
obtain in real life. We demonstrate the efficacy of our model on the real life
StanfordORB benchmark where, given few input views, we achieve reconstruction
quality significantly superior to existing feedforward reconstruction networks,
and comparable to significantly slower per-scene optimization methods.
| new_dataset | 0.957118 |
2503.08384 | Susu Sun | Susu Sun, Dominique van Midden, Geert Litjens, Christian F.
Baumgartner | Prototype-Based Multiple Instance Learning for Gigapixel Whole Slide
Image Classification | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | Multiple Instance Learning (MIL) methods have succeeded remarkably in
histopathology whole slide image (WSI) analysis. However, most MIL models only
offer attention-based explanations that do not faithfully capture the model's
decision mechanism and do not allow human-model interaction. To address these
limitations, we introduce ProtoMIL, an inherently interpretable MIL model for
WSI analysis that offers user-friendly explanations and supports human
intervention. Our approach employs a sparse autoencoder to discover
human-interpretable concepts from the image feature space, which are then used
to train ProtoMIL. The model represents predictions as linear combinations of
concepts, making the decision process transparent. Furthermore, ProtoMIL allows
users to perform model interventions by altering the input concepts.
Experiments on two widely used pathology datasets demonstrate that ProtoMIL
achieves a classification performance comparable to state-of-the-art MIL models
while offering intuitively understandable explanations. Moreover, we
demonstrate that our method can eliminate reliance on diagnostically irrelevant
information via human intervention, guiding the model toward being right for
the right reason. Code will be publicly available at
https://github.com/ss-sun/ProtoMIL.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 12:44:03 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Sun",
"Susu",
""
],
[
"van Midden",
"Dominique",
""
],
[
"Litjens",
"Geert",
""
],
[
"Baumgartner",
"Christian F.",
""
]
]
| TITLE: Prototype-Based Multiple Instance Learning for Gigapixel Whole Slide
Image Classification
ABSTRACT: Multiple Instance Learning (MIL) methods have succeeded remarkably in
histopathology whole slide image (WSI) analysis. However, most MIL models only
offer attention-based explanations that do not faithfully capture the model's
decision mechanism and do not allow human-model interaction. To address these
limitations, we introduce ProtoMIL, an inherently interpretable MIL model for
WSI analysis that offers user-friendly explanations and supports human
intervention. Our approach employs a sparse autoencoder to discover
human-interpretable concepts from the image feature space, which are then used
to train ProtoMIL. The model represents predictions as linear combinations of
concepts, making the decision process transparent. Furthermore, ProtoMIL allows
users to perform model interventions by altering the input concepts.
Experiments on two widely used pathology datasets demonstrate that ProtoMIL
achieves a classification performance comparable to state-of-the-art MIL models
while offering intuitively understandable explanations. Moreover, we
demonstrate that our method can eliminate reliance on diagnostically irrelevant
information via human intervention, guiding the model toward being right for
the right reason. Code will be publicly available at
https://github.com/ss-sun/ProtoMIL.
| no_new_dataset | 0.944791 |
2503.08388 | Wa\"el Doulazmi | Valentin Charraut and Thomas Tournaire and Wa\"el Doulazmi and
Thibault Buhet | V-Max: Making RL practical for Autonomous Driving | null | null | null | null | cs.LG cs.AI cs.RO | http://creativecommons.org/licenses/by/4.0/ | Learning-based decision-making has the potential to enable generalizable
Autonomous Driving (AD) policies, reducing the engineering overhead of
rule-based approaches. Imitation Learning (IL) remains the dominant paradigm,
benefiting from large-scale human demonstration datasets, but it suffers from
inherent limitations such as distribution shift and imitation gaps.
Reinforcement Learning (RL) presents a promising alternative, yet its adoption
in AD remains limited due to the lack of standardized and efficient research
frameworks. To this end, we introduce V-Max, an open research framework
providing all the necessary tools to make RL practical for AD. V-Max is built
on Waymax, a hardware-accelerated AD simulator designed for large-scale
experimentation. We extend it using ScenarioNet's approach, enabling the fast
simulation of diverse AD datasets. V-Max integrates a set of observation and
reward functions, transformer-based encoders, and training pipelines.
Additionally, it includes adversarial evaluation settings and an extensive set
of evaluation metrics. Through a large-scale benchmark, we analyze how network
architectures, observation functions, training data, and reward shaping impact
RL performance.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 12:53:24 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Charraut",
"Valentin",
""
],
[
"Tournaire",
"Thomas",
""
],
[
"Doulazmi",
"Waël",
""
],
[
"Buhet",
"Thibault",
""
]
]
| TITLE: V-Max: Making RL practical for Autonomous Driving
ABSTRACT: Learning-based decision-making has the potential to enable generalizable
Autonomous Driving (AD) policies, reducing the engineering overhead of
rule-based approaches. Imitation Learning (IL) remains the dominant paradigm,
benefiting from large-scale human demonstration datasets, but it suffers from
inherent limitations such as distribution shift and imitation gaps.
Reinforcement Learning (RL) presents a promising alternative, yet its adoption
in AD remains limited due to the lack of standardized and efficient research
frameworks. To this end, we introduce V-Max, an open research framework
providing all the necessary tools to make RL practical for AD. V-Max is built
on Waymax, a hardware-accelerated AD simulator designed for large-scale
experimentation. We extend it using ScenarioNet's approach, enabling the fast
simulation of diverse AD datasets. V-Max integrates a set of observation and
reward functions, transformer-based encoders, and training pipelines.
Additionally, it includes adversarial evaluation settings and an extensive set
of evaluation metrics. Through a large-scale benchmark, we analyze how network
architectures, observation functions, training data, and reward shaping impact
RL performance.
| no_new_dataset | 0.939192 |
2503.08410 | Marcos Cirne | Marcos Cirne, Hannah Menke, Alhasan Abdellatif, Julien Maes, Florian
Doster, Ahmed H. Elsheikh | A Deep-Learning Iterative Stacked Approach for Prediction of Reactive
Dissolution in Porous Media | 24 pages, 16 figures | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Simulating reactive dissolution of solid minerals in porous media has many
subsurface applications, including carbon capture and storage (CCS), geothermal
systems and oil & gas recovery. As traditional direct numerical simulators are
computationally expensive, it is of paramount importance to develop faster and
more efficient alternatives. Deep-learning-based solutions, most of them built
upon convolutional neural networks (CNNs), have been recently designed to
tackle this problem. However, these solutions were limited to approximating one
field over the domain (e.g. velocity field). In this manuscript, we present a
novel deep learning approach that incorporates both temporal and spatial
information to predict the future states of the dissolution process at a fixed
time-step horizon, given a sequence of input states. The overall performance,
in terms of speed and prediction accuracy, is demonstrated on a numerical
simulation dataset, comparing its prediction results against state-of-the-art
approaches, also achieving a speedup around $10^4$ over traditional numerical
simulators.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 13:18:03 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Cirne",
"Marcos",
""
],
[
"Menke",
"Hannah",
""
],
[
"Abdellatif",
"Alhasan",
""
],
[
"Maes",
"Julien",
""
],
[
"Doster",
"Florian",
""
],
[
"Elsheikh",
"Ahmed H.",
""
]
]
| TITLE: A Deep-Learning Iterative Stacked Approach for Prediction of Reactive
Dissolution in Porous Media
ABSTRACT: Simulating reactive dissolution of solid minerals in porous media has many
subsurface applications, including carbon capture and storage (CCS), geothermal
systems and oil & gas recovery. As traditional direct numerical simulators are
computationally expensive, it is of paramount importance to develop faster and
more efficient alternatives. Deep-learning-based solutions, most of them built
upon convolutional neural networks (CNNs), have been recently designed to
tackle this problem. However, these solutions were limited to approximating one
field over the domain (e.g. velocity field). In this manuscript, we present a
novel deep learning approach that incorporates both temporal and spatial
information to predict the future states of the dissolution process at a fixed
time-step horizon, given a sequence of input states. The overall performance,
in terms of speed and prediction accuracy, is demonstrated on a numerical
simulation dataset, comparing its prediction results against state-of-the-art
approaches, also achieving a speedup around $10^4$ over traditional numerical
simulators.
| no_new_dataset | 0.943919 |
2503.08417 | Kwan Yun | Kwan Yun, Seokhyeon Hong, Chaelin Kim, Junyong Noh | AnyMoLe: Any Character Motion In-betweening Leveraging Video Diffusion
Models | 11 pages, 10 figures, CVPR 2025 | null | null | null | cs.GR cs.AI cs.CV cs.LG cs.MM | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Despite recent advancements in learning-based motion in-betweening, a key
limitation has been overlooked: the requirement for character-specific
datasets. In this work, we introduce AnyMoLe, a novel method that addresses
this limitation by leveraging video diffusion models to generate motion
in-between frames for arbitrary characters without external data. Our approach
employs a two-stage frame generation process to enhance contextual
understanding. Furthermore, to bridge the domain gap between real-world and
rendered character animations, we introduce ICAdapt, a fine-tuning technique
for video diffusion models. Additionally, we propose a ``motion-video
mimicking'' optimization technique, enabling seamless motion generation for
characters with arbitrary joint structures using 2D and 3D-aware features.
AnyMoLe significantly reduces data dependency while generating smooth and
realistic transitions, making it applicable to a wide range of motion
in-betweening tasks.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 13:28:59 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Yun",
"Kwan",
""
],
[
"Hong",
"Seokhyeon",
""
],
[
"Kim",
"Chaelin",
""
],
[
"Noh",
"Junyong",
""
]
]
| TITLE: AnyMoLe: Any Character Motion In-betweening Leveraging Video Diffusion
Models
ABSTRACT: Despite recent advancements in learning-based motion in-betweening, a key
limitation has been overlooked: the requirement for character-specific
datasets. In this work, we introduce AnyMoLe, a novel method that addresses
this limitation by leveraging video diffusion models to generate motion
in-between frames for arbitrary characters without external data. Our approach
employs a two-stage frame generation process to enhance contextual
understanding. Furthermore, to bridge the domain gap between real-world and
rendered character animations, we introduce ICAdapt, a fine-tuning technique
for video diffusion models. Additionally, we propose a ``motion-video
mimicking'' optimization technique, enabling seamless motion generation for
characters with arbitrary joint structures using 2D and 3D-aware features.
AnyMoLe significantly reduces data dependency while generating smooth and
realistic transitions, making it applicable to a wide range of motion
in-betweening tasks.
| no_new_dataset | 0.943867 |
2503.08420 | Yihang Wu | Ahmad Chaddad, Yan Hu, Yihang Wu, Binbin Wen, Reem Kateb | Generalizable and Explainable Deep Learning for Medical Image Computing:
An Overview | Published in Current Opinion in Biomedical Engineering | null | 10.1016/j.cobme.2024.100567 | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Objective. This paper presents an overview of generalizable and explainable
artificial intelligence (XAI) in deep learning (DL) for medical imaging, aimed
at addressing the urgent need for transparency and explainability in clinical
applications.
Methodology. We propose to use four CNNs in three medical datasets (brain
tumor, skin cancer, and chest x-ray) for medical image classification tasks. In
addition, we perform paired t-tests to show the significance of the differences
observed between different methods. Furthermore, we propose to combine ResNet50
with five common XAI techniques to obtain explainable results for model
prediction, aiming at improving model transparency. We also involve a
quantitative metric (confidence increase) to evaluate the usefulness of XAI
techniques.
Key findings. The experimental results indicate that ResNet50 can achieve
feasible accuracy and F1 score in all datasets (e.g., 86.31\% accuracy in skin
cancer). Furthermore, the findings show that while certain XAI methods, such as
XgradCAM, effectively highlight relevant abnormal regions in medical images,
others, like EigenGradCAM, may perform less effectively in specific scenarios.
In addition, XgradCAM indicates higher confidence increase (e.g., 0.12 in
glioma tumor) compared to GradCAM++ (0.09) and LayerCAM (0.08).
Implications. Based on the experimental results and recent advancements, we
outline future research directions to enhance the robustness and
generalizability of DL models in the field of biomedical imaging.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 13:31:09 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Chaddad",
"Ahmad",
""
],
[
"Hu",
"Yan",
""
],
[
"Wu",
"Yihang",
""
],
[
"Wen",
"Binbin",
""
],
[
"Kateb",
"Reem",
""
]
]
| TITLE: Generalizable and Explainable Deep Learning for Medical Image Computing:
An Overview
ABSTRACT: Objective. This paper presents an overview of generalizable and explainable
artificial intelligence (XAI) in deep learning (DL) for medical imaging, aimed
at addressing the urgent need for transparency and explainability in clinical
applications.
Methodology. We propose to use four CNNs in three medical datasets (brain
tumor, skin cancer, and chest x-ray) for medical image classification tasks. In
addition, we perform paired t-tests to show the significance of the differences
observed between different methods. Furthermore, we propose to combine ResNet50
with five common XAI techniques to obtain explainable results for model
prediction, aiming at improving model transparency. We also involve a
quantitative metric (confidence increase) to evaluate the usefulness of XAI
techniques.
Key findings. The experimental results indicate that ResNet50 can achieve
feasible accuracy and F1 score in all datasets (e.g., 86.31\% accuracy in skin
cancer). Furthermore, the findings show that while certain XAI methods, such as
XgradCAM, effectively highlight relevant abnormal regions in medical images,
others, like EigenGradCAM, may perform less effectively in specific scenarios.
In addition, XgradCAM indicates higher confidence increase (e.g., 0.12 in
glioma tumor) compared to GradCAM++ (0.09) and LayerCAM (0.08).
Implications. Based on the experimental results and recent advancements, we
outline future research directions to enhance the robustness and
generalizability of DL models in the field of biomedical imaging.
| no_new_dataset | 0.949529 |
2503.08437 | Shankar Gangisetty | Shankar Gangisetty, Abdul Wasi, Shyam Nandan Rai, C. V. Jawahar, Sajay
Raj, Manish Prajapati, Ayesha Choudhary, Aaryadev Chandra, Dev Chandan,
Shireen Chand, Suvaditya Mukherjee | ICPR 2024 Competition on Rider Intention Prediction | null | null | 10.1007/978-3-031-80139-6_3 | null | cs.CV cs.AI cs.HC cs.RO | http://creativecommons.org/licenses/by/4.0/ | The recent surge in the vehicle market has led to an alarming increase in
road accidents. This underscores the critical importance of enhancing road
safety measures, particularly for vulnerable road users like motorcyclists.
Hence, we introduce the rider intention prediction (RIP) competition that aims
to address challenges in rider safety by proactively predicting maneuvers
before they occur, thereby strengthening rider safety. This capability enables
the riders to react to the potential incorrect maneuvers flagged by advanced
driver assistance systems (ADAS). We collect a new dataset, namely, rider
action anticipation dataset (RAAD) for the competition consisting of two tasks:
single-view RIP and multi-view RIP. The dataset incorporates a spectrum of
traffic conditions and challenging navigational maneuvers on roads with varying
lighting conditions. For the competition, we received seventy-five
registrations and five team submissions for inference of which we compared the
methods of the top three performing teams on both the RIP tasks: one
state-space model (Mamba2) and two learning-based approaches (SVM and
CNN-LSTM). The results indicate that the state-space model outperformed the
other methods across the entire dataset, providing a balanced performance
across maneuver classes. The SVM-based RIP method showed the second-best
performance when using random sampling and SMOTE. However, the CNN-LSTM method
underperformed, primarily due to class imbalance issues, particularly
struggling with minority classes. This paper details the proposed RAAD dataset
and provides a summary of the submissions for the RIP 2024 competition.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 13:50:37 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Gangisetty",
"Shankar",
""
],
[
"Wasi",
"Abdul",
""
],
[
"Rai",
"Shyam Nandan",
""
],
[
"Jawahar",
"C. V.",
""
],
[
"Raj",
"Sajay",
""
],
[
"Prajapati",
"Manish",
""
],
[
"Choudhary",
"Ayesha",
""
],
[
"Chandra",
"Aaryadev",
""
],
[
"Chandan",
"Dev",
""
],
[
"Chand",
"Shireen",
""
],
[
"Mukherjee",
"Suvaditya",
""
]
]
| TITLE: ICPR 2024 Competition on Rider Intention Prediction
ABSTRACT: The recent surge in the vehicle market has led to an alarming increase in
road accidents. This underscores the critical importance of enhancing road
safety measures, particularly for vulnerable road users like motorcyclists.
Hence, we introduce the rider intention prediction (RIP) competition that aims
to address challenges in rider safety by proactively predicting maneuvers
before they occur, thereby strengthening rider safety. This capability enables
the riders to react to the potential incorrect maneuvers flagged by advanced
driver assistance systems (ADAS). We collect a new dataset, namely, rider
action anticipation dataset (RAAD) for the competition consisting of two tasks:
single-view RIP and multi-view RIP. The dataset incorporates a spectrum of
traffic conditions and challenging navigational maneuvers on roads with varying
lighting conditions. For the competition, we received seventy-five
registrations and five team submissions for inference of which we compared the
methods of the top three performing teams on both the RIP tasks: one
state-space model (Mamba2) and two learning-based approaches (SVM and
CNN-LSTM). The results indicate that the state-space model outperformed the
other methods across the entire dataset, providing a balanced performance
across maneuver classes. The SVM-based RIP method showed the second-best
performance when using random sampling and SMOTE. However, the CNN-LSTM method
underperformed, primarily due to class imbalance issues, particularly
struggling with minority classes. This paper details the proposed RAAD dataset
and provides a summary of the submissions for the RIP 2024 competition.
| new_dataset | 0.971564 |
2503.08461 | Jianian Zhu | Jianian Zhu, Hang Wu, Haojie Wang, Yinghui Li, Biao Hou, Ruixuan Li,
Jidong Zhai | FastCache: Optimizing Multimodal LLM Serving through Lightweight
KV-Cache Compression Framework | 14 pages, 14 figures | null | null | null | cs.MM cs.DC | http://creativecommons.org/licenses/by/4.0/ | Multi-modal Large Language Models (MLLMs) serving systems commonly employ
KV-cache compression to reduce memory footprint. However, existing compression
methods introduce significant processing overhead and queuing delays,
particularly in concurrent serving scenarios. We present \texttt{FastCache}, a
novel serving framework that effectively addresses these challenges through two
key innovations: (1) a dynamic batching strategy that optimizes request
scheduling across prefill, compression, and decode stages, and (2) an efficient
KV-cache memory pool mechanism that eliminates memory fragmentation while
maintaining high GPU utilization. Our comprehensive experiments on the GQA and
MileBench datasets demonstrate that \texttt{FastCache} achieves up to
19.3$\times$ reduction in Time-To-First-Token (TTFT) and 12.1$\times$
improvement in throughput compared to state-of-the-art baselines. The system
maintains stable performance under high-concurrency scenarios (up to 40 req/s)
while reducing average memory consumption by 20\%. These results establish
\texttt{FastCache} as an efficient solution for real-world LLM serving systems
with KV-cache compression.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 14:10:58 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Zhu",
"Jianian",
""
],
[
"Wu",
"Hang",
""
],
[
"Wang",
"Haojie",
""
],
[
"Li",
"Yinghui",
""
],
[
"Hou",
"Biao",
""
],
[
"Li",
"Ruixuan",
""
],
[
"Zhai",
"Jidong",
""
]
]
| TITLE: FastCache: Optimizing Multimodal LLM Serving through Lightweight
KV-Cache Compression Framework
ABSTRACT: Multi-modal Large Language Models (MLLMs) serving systems commonly employ
KV-cache compression to reduce memory footprint. However, existing compression
methods introduce significant processing overhead and queuing delays,
particularly in concurrent serving scenarios. We present \texttt{FastCache}, a
novel serving framework that effectively addresses these challenges through two
key innovations: (1) a dynamic batching strategy that optimizes request
scheduling across prefill, compression, and decode stages, and (2) an efficient
KV-cache memory pool mechanism that eliminates memory fragmentation while
maintaining high GPU utilization. Our comprehensive experiments on the GQA and
MileBench datasets demonstrate that \texttt{FastCache} achieves up to
19.3$\times$ reduction in Time-To-First-Token (TTFT) and 12.1$\times$
improvement in throughput compared to state-of-the-art baselines. The system
maintains stable performance under high-concurrency scenarios (up to 40 req/s)
while reducing average memory consumption by 20\%. These results establish
\texttt{FastCache} as an efficient solution for real-world LLM serving systems
with KV-cache compression.
| no_new_dataset | 0.942135 |
2503.08463 | Junyoung Kim | Junyoung Kim, Madhulika Balakumar, Kenneth Ross | A Data Aggregation Visualization System supported by
Processing-in-Memory | 13 pages, 11 figures | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Data visualization of aggregation queries is one of the most common ways of
doing data exploration and data science as it can help identify correlations
and patterns in the data. We propose DIVAN, a system that automatically
normalizes the one-dimensional axes by frequency to generate large numbers of
two-dimensional visualizations. DIVAN normalizes the input data via binning to
allocate more pixels to data values that appear more frequently in the dataset.
DIVAN can utilize either CPUs or Processing-in-Memory (PIM) architectures to
quickly calculate aggregates to support the visualizations. On real world
datasets, we show that DIVAN generates visualizations that highlight patterns
and correlations, some expected and some unexpected. By using PIM, we can
calculate aggregates 45%-64% faster than modern CPUs on large datasets. For use
cases with 100 million rows and 32 columns, our system is able to compute 4,960
aggregates (each of size 128x128x128) in about a minute.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 14:12:46 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Kim",
"Junyoung",
""
],
[
"Balakumar",
"Madhulika",
""
],
[
"Ross",
"Kenneth",
""
]
]
| TITLE: A Data Aggregation Visualization System supported by
Processing-in-Memory
ABSTRACT: Data visualization of aggregation queries is one of the most common ways of
doing data exploration and data science as it can help identify correlations
and patterns in the data. We propose DIVAN, a system that automatically
normalizes the one-dimensional axes by frequency to generate large numbers of
two-dimensional visualizations. DIVAN normalizes the input data via binning to
allocate more pixels to data values that appear more frequently in the dataset.
DIVAN can utilize either CPUs or Processing-in-Memory (PIM) architectures to
quickly calculate aggregates to support the visualizations. On real world
datasets, we show that DIVAN generates visualizations that highlight patterns
and correlations, some expected and some unexpected. By using PIM, we can
calculate aggregates 45%-64% faster than modern CPUs on large datasets. For use
cases with 100 million rows and 32 columns, our system is able to compute 4,960
aggregates (each of size 128x128x128) in about a minute.
| no_new_dataset | 0.948058 |
2503.08471 | Zhuoguang Chen | Zhuoguang Chen, Kenan Li, Xiuyu Yang, Tao Jiang, Yiming Li, Hang Zhao | TrackOcc: Camera-based 4D Panoptic Occupancy Tracking | Accepted at ICRA 2025 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Comprehensive and consistent dynamic scene understanding from camera input is
essential for advanced autonomous systems. Traditional camera-based perception
tasks like 3D object tracking and semantic occupancy prediction lack either
spatial comprehensiveness or temporal consistency. In this work, we introduce a
brand-new task, Camera-based 4D Panoptic Occupancy Tracking, which
simultaneously addresses panoptic occupancy segmentation and object tracking
from camera-only input. Furthermore, we propose TrackOcc, a cutting-edge
approach that processes image inputs in a streaming, end-to-end manner with 4D
panoptic queries to address the proposed task. Leveraging the
localization-aware loss, TrackOcc enhances the accuracy of 4D panoptic
occupancy tracking without bells and whistles. Experimental results demonstrate
that our method achieves state-of-the-art performance on the Waymo dataset. The
source code will be released at https://github.com/Tsinghua-MARS-Lab/TrackOcc.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 14:17:06 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Chen",
"Zhuoguang",
""
],
[
"Li",
"Kenan",
""
],
[
"Yang",
"Xiuyu",
""
],
[
"Jiang",
"Tao",
""
],
[
"Li",
"Yiming",
""
],
[
"Zhao",
"Hang",
""
]
]
| TITLE: TrackOcc: Camera-based 4D Panoptic Occupancy Tracking
ABSTRACT: Comprehensive and consistent dynamic scene understanding from camera input is
essential for advanced autonomous systems. Traditional camera-based perception
tasks like 3D object tracking and semantic occupancy prediction lack either
spatial comprehensiveness or temporal consistency. In this work, we introduce a
brand-new task, Camera-based 4D Panoptic Occupancy Tracking, which
simultaneously addresses panoptic occupancy segmentation and object tracking
from camera-only input. Furthermore, we propose TrackOcc, a cutting-edge
approach that processes image inputs in a streaming, end-to-end manner with 4D
panoptic queries to address the proposed task. Leveraging the
localization-aware loss, TrackOcc enhances the accuracy of 4D panoptic
occupancy tracking without bells and whistles. Experimental results demonstrate
that our method achieves state-of-the-art performance on the Waymo dataset. The
source code will be released at https://github.com/Tsinghua-MARS-Lab/TrackOcc.
| no_new_dataset | 0.939248 |
2503.08472 | Hao Jiang | Hao Jiang, Yixing Xu, Pradeep Varakantham | Optimizing Ride-Pooling Operations with Extended Pickup and Drop-Off
Flexibility | null | null | null | null | cs.RO cs.AI | http://creativecommons.org/licenses/by/4.0/ | The Ride-Pool Matching Problem (RMP) is central to on-demand ride-pooling
services, where vehicles must be matched with multiple requests while adhering
to service constraints such as pickup delays, detour limits, and vehicle
capacity. Most existing RMP solutions assume passengers are picked up and
dropped off at their original locations, neglecting the potential for
passengers to walk to nearby spots to meet vehicles. This assumption restricts
the optimization potential in ride-pooling operations. In this paper, we
propose a novel matching method that incorporates extended pickup and drop-off
areas for passengers. We first design a tree-based approach to efficiently
generate feasible matches between passengers and vehicles. Next, we optimize
vehicle routes to cover all designated pickup and drop-off locations while
minimizing total travel distance. Finally, we employ dynamic assignment
strategies to achieve optimal matching outcomes. Experiments on city-scale taxi
datasets demonstrate that our method improves the number of served requests by
up to 13\% and average travel distance by up to 21\% compared to leading
existing solutions, underscoring the potential of leveraging passenger mobility
to significantly enhance ride-pooling service efficiency.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 14:17:30 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Jiang",
"Hao",
""
],
[
"Xu",
"Yixing",
""
],
[
"Varakantham",
"Pradeep",
""
]
]
| TITLE: Optimizing Ride-Pooling Operations with Extended Pickup and Drop-Off
Flexibility
ABSTRACT: The Ride-Pool Matching Problem (RMP) is central to on-demand ride-pooling
services, where vehicles must be matched with multiple requests while adhering
to service constraints such as pickup delays, detour limits, and vehicle
capacity. Most existing RMP solutions assume passengers are picked up and
dropped off at their original locations, neglecting the potential for
passengers to walk to nearby spots to meet vehicles. This assumption restricts
the optimization potential in ride-pooling operations. In this paper, we
propose a novel matching method that incorporates extended pickup and drop-off
areas for passengers. We first design a tree-based approach to efficiently
generate feasible matches between passengers and vehicles. Next, we optimize
vehicle routes to cover all designated pickup and drop-off locations while
minimizing total travel distance. Finally, we employ dynamic assignment
strategies to achieve optimal matching outcomes. Experiments on city-scale taxi
datasets demonstrate that our method improves the number of served requests by
up to 13\% and average travel distance by up to 21\% compared to leading
existing solutions, underscoring the potential of leveraging passenger mobility
to significantly enhance ride-pooling service efficiency.
| no_new_dataset | 0.945851 |
2503.08474 | Martin Alexander B\"uchner | Tim Steinke, Martin B\"uchner, Niclas V\"odisch, and Abhinav Valada | Collaborative Dynamic 3D Scene Graphs for Open-Vocabulary Urban Scene
Understanding | null | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Mapping and scene representation are fundamental to reliable planning and
navigation in mobile robots. While purely geometric maps using voxel grids
allow for general navigation, obtaining up-to-date spatial and semantically
rich representations that scale to dynamic large-scale environments remains
challenging. In this work, we present CURB-OSG, an open-vocabulary dynamic 3D
scene graph engine that generates hierarchical decompositions of urban driving
scenes via multi-agent collaboration. By fusing the camera and LiDAR
observations from multiple perceiving agents with unknown initial poses, our
approach generates more accurate maps compared to a single agent while
constructing a unified open-vocabulary semantic hierarchy of the scene. Unlike
previous methods that rely on ground truth agent poses or are evaluated purely
in simulation, CURB-OSG alleviates these constraints. We evaluate the
capabilities of CURB-OSG on real-world multi-agent sensor data obtained from
multiple sessions of the Oxford Radar RobotCar dataset. We demonstrate improved
mapping and object prediction accuracy through multi-agent collaboration as
well as evaluate the environment partitioning capabilities of the proposed
approach. To foster further research, we release our code and supplementary
material at https://ov-curb.cs.uni-freiburg.de.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 14:21:59 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Steinke",
"Tim",
""
],
[
"Büchner",
"Martin",
""
],
[
"Vödisch",
"Niclas",
""
],
[
"Valada",
"Abhinav",
""
]
]
| TITLE: Collaborative Dynamic 3D Scene Graphs for Open-Vocabulary Urban Scene
Understanding
ABSTRACT: Mapping and scene representation are fundamental to reliable planning and
navigation in mobile robots. While purely geometric maps using voxel grids
allow for general navigation, obtaining up-to-date spatial and semantically
rich representations that scale to dynamic large-scale environments remains
challenging. In this work, we present CURB-OSG, an open-vocabulary dynamic 3D
scene graph engine that generates hierarchical decompositions of urban driving
scenes via multi-agent collaboration. By fusing the camera and LiDAR
observations from multiple perceiving agents with unknown initial poses, our
approach generates more accurate maps compared to a single agent while
constructing a unified open-vocabulary semantic hierarchy of the scene. Unlike
previous methods that rely on ground truth agent poses or are evaluated purely
in simulation, CURB-OSG alleviates these constraints. We evaluate the
capabilities of CURB-OSG on real-world multi-agent sensor data obtained from
multiple sessions of the Oxford Radar RobotCar dataset. We demonstrate improved
mapping and object prediction accuracy through multi-agent collaboration as
well as evaluate the environment partitioning capabilities of the proposed
approach. To foster further research, we release our code and supplementary
material at https://ov-curb.cs.uni-freiburg.de.
| no_new_dataset | 0.944485 |
2503.08482 | Pouya Shaeri | Pouya Shaeri, Saud AlKhaled, Ariane Middel | A Multimodal Physics-Informed Neural Network Approach for Mean Radiant
Temperature Modeling | null | null | null | null | cs.CV cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Outdoor thermal comfort is a critical determinant of urban livability,
particularly in hot desert climates where extreme heat poses challenges to
public health, energy consumption, and urban planning. Mean Radiant Temperature
($T_{mrt}$) is a key parameter for evaluating outdoor thermal comfort,
especially in urban environments where radiation dynamics significantly impact
human thermal exposure. Traditional methods of estimating $T_{mrt}$ rely on
field measurements and computational simulations, both of which are resource
intensive. This study introduces a Physics-Informed Neural Network (PINN)
approach that integrates shortwave and longwave radiation modeling with deep
learning techniques. By leveraging a multimodal dataset that includes
meteorological data, built environment characteristics, and fisheye
image-derived shading information, our model enhances predictive accuracy while
maintaining physical consistency. Our experimental results demonstrate that the
proposed PINN framework outperforms conventional deep learning models, with the
best-performing configurations achieving an RMSE of 3.50 and an $R^2$ of 0.88.
This approach highlights the potential of physics-informed machine learning in
bridging the gap between computational modeling and real-world applications,
offering a scalable and interpretable solution for urban thermal comfort
assessments.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 14:36:08 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Shaeri",
"Pouya",
""
],
[
"AlKhaled",
"Saud",
""
],
[
"Middel",
"Ariane",
""
]
]
| TITLE: A Multimodal Physics-Informed Neural Network Approach for Mean Radiant
Temperature Modeling
ABSTRACT: Outdoor thermal comfort is a critical determinant of urban livability,
particularly in hot desert climates where extreme heat poses challenges to
public health, energy consumption, and urban planning. Mean Radiant Temperature
($T_{mrt}$) is a key parameter for evaluating outdoor thermal comfort,
especially in urban environments where radiation dynamics significantly impact
human thermal exposure. Traditional methods of estimating $T_{mrt}$ rely on
field measurements and computational simulations, both of which are resource
intensive. This study introduces a Physics-Informed Neural Network (PINN)
approach that integrates shortwave and longwave radiation modeling with deep
learning techniques. By leveraging a multimodal dataset that includes
meteorological data, built environment characteristics, and fisheye
image-derived shading information, our model enhances predictive accuracy while
maintaining physical consistency. Our experimental results demonstrate that the
proposed PINN framework outperforms conventional deep learning models, with the
best-performing configurations achieving an RMSE of 3.50 and an $R^2$ of 0.88.
This approach highlights the potential of physics-informed machine learning in
bridging the gap between computational modeling and real-world applications,
offering a scalable and interpretable solution for urban thermal comfort
assessments.
| no_new_dataset | 0.946745 |
2503.08483 | Abhishek Saroha | Nhat Phuong Anh Vu, Abhishek Saroha, Or Litany, Daniel Cremers | GAS-NeRF: Geometry-Aware Stylization of Dynamic Radiance Fields | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Current 3D stylization techniques primarily focus on static scenes, while our
world is inherently dynamic, filled with moving objects and changing
environments. Existing style transfer methods primarily target appearance --
such as color and texture transformation -- but often neglect the geometric
characteristics of the style image, which are crucial for achieving a complete
and coherent stylization effect. To overcome these shortcomings, we propose
GAS-NeRF, a novel approach for joint appearance and geometry stylization in
dynamic Radiance Fields. Our method leverages depth maps to extract and
transfer geometric details into the radiance field, followed by appearance
transfer. Experimental results on synthetic and real-world datasets demonstrate
that our approach significantly enhances the stylization quality while
maintaining temporal coherence in dynamic scenes.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 14:37:06 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Vu",
"Nhat Phuong Anh",
""
],
[
"Saroha",
"Abhishek",
""
],
[
"Litany",
"Or",
""
],
[
"Cremers",
"Daniel",
""
]
]
| TITLE: GAS-NeRF: Geometry-Aware Stylization of Dynamic Radiance Fields
ABSTRACT: Current 3D stylization techniques primarily focus on static scenes, while our
world is inherently dynamic, filled with moving objects and changing
environments. Existing style transfer methods primarily target appearance --
such as color and texture transformation -- but often neglect the geometric
characteristics of the style image, which are crucial for achieving a complete
and coherent stylization effect. To overcome these shortcomings, we propose
GAS-NeRF, a novel approach for joint appearance and geometry stylization in
dynamic Radiance Fields. Our method leverages depth maps to extract and
transfer geometric details into the radiance field, followed by appearance
transfer. Experimental results on synthetic and real-world datasets demonstrate
that our approach significantly enhances the stylization quality while
maintaining temporal coherence in dynamic scenes.
| no_new_dataset | 0.954393 |
2503.08495 | Han Cao | Han Cao, Lingwei Wei, Wei Zhou, Songlin Hu | Enhancing Multi-Hop Fact Verification with Structured
Knowledge-Augmented Large Language Models | Accepted by AAAI 2025 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The rapid development of social platforms exacerbates the dissemination of
misinformation, which stimulates the research in fact verification. Recent
studies tend to leverage semantic features to solve this problem as a
single-hop task. However, the process of verifying a claim requires several
pieces of evidence with complicated inner logic and relations to verify the
given claim in real-world situations. Recent studies attempt to improve both
understanding and reasoning abilities to enhance the performance, but they
overlook the crucial relations between entities that benefit models to
understand better and facilitate the prediction. To emphasize the significance
of relations, we resort to Large Language Models (LLMs) considering their
excellent understanding ability. Instead of other methods using LLMs as the
predictor, we take them as relation extractors, for they do better in
understanding rather than reasoning according to the experimental results.
Thus, to solve the challenges above, we propose a novel Structured
Knowledge-Augmented LLM-based Network (LLM-SKAN) for multi-hop fact
verification. Specifically, we utilize an LLM-driven Knowledge Extractor to
capture fine-grained information, including entities and their complicated
relations. Besides, we leverage a Knowledge-Augmented Relation Graph Fusion
module to interact with each node and learn better claim-evidence
representations comprehensively. The experimental results on four common-used
datasets demonstrate the effectiveness and superiority of our model.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 14:47:24 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Cao",
"Han",
""
],
[
"Wei",
"Lingwei",
""
],
[
"Zhou",
"Wei",
""
],
[
"Hu",
"Songlin",
""
]
]
| TITLE: Enhancing Multi-Hop Fact Verification with Structured
Knowledge-Augmented Large Language Models
ABSTRACT: The rapid development of social platforms exacerbates the dissemination of
misinformation, which stimulates the research in fact verification. Recent
studies tend to leverage semantic features to solve this problem as a
single-hop task. However, the process of verifying a claim requires several
pieces of evidence with complicated inner logic and relations to verify the
given claim in real-world situations. Recent studies attempt to improve both
understanding and reasoning abilities to enhance the performance, but they
overlook the crucial relations between entities that benefit models to
understand better and facilitate the prediction. To emphasize the significance
of relations, we resort to Large Language Models (LLMs) considering their
excellent understanding ability. Instead of other methods using LLMs as the
predictor, we take them as relation extractors, for they do better in
understanding rather than reasoning according to the experimental results.
Thus, to solve the challenges above, we propose a novel Structured
Knowledge-Augmented LLM-based Network (LLM-SKAN) for multi-hop fact
verification. Specifically, we utilize an LLM-driven Knowledge Extractor to
capture fine-grained information, including entities and their complicated
relations. Besides, we leverage a Knowledge-Augmented Relation Graph Fusion
module to interact with each node and learn better claim-evidence
representations comprehensively. The experimental results on four common-used
datasets demonstrate the effectiveness and superiority of our model.
| no_new_dataset | 0.949716 |
2503.08496 | Henry Senior | Henry Senior, Luca Rossi, Gregory Slabaugh, Shanxin Yuan | SuperCap: Multi-resolution Superpixel-based Image Captioning | 12 pages, 4 figures | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | It has been a longstanding goal within image captioning to move beyond a
dependence on object detection. We investigate using superpixels coupled with
Vision Language Models (VLMs) to bridge the gap between detector-based
captioning architectures and those that solely pretrain on large datasets. Our
novel superpixel approach ensures that the model receives object-like features
whilst the use of VLMs provides our model with open set object understanding.
Furthermore, we extend our architecture to make use of multi-resolution inputs,
allowing our model to view images in different levels of detail, and use an
attention mechanism to determine which parts are most relevant to the caption.
We demonstrate our model's performance with multiple VLMs and through a range
of ablations detailing the impact of different architectural choices. Our full
model achieves a competitive CIDEr score of $136.9$ on the COCO Karpathy split.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 14:47:46 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Senior",
"Henry",
""
],
[
"Rossi",
"Luca",
""
],
[
"Slabaugh",
"Gregory",
""
],
[
"Yuan",
"Shanxin",
""
]
]
| TITLE: SuperCap: Multi-resolution Superpixel-based Image Captioning
ABSTRACT: It has been a longstanding goal within image captioning to move beyond a
dependence on object detection. We investigate using superpixels coupled with
Vision Language Models (VLMs) to bridge the gap between detector-based
captioning architectures and those that solely pretrain on large datasets. Our
novel superpixel approach ensures that the model receives object-like features
whilst the use of VLMs provides our model with open set object understanding.
Furthermore, we extend our architecture to make use of multi-resolution inputs,
allowing our model to view images in different levels of detail, and use an
attention mechanism to determine which parts are most relevant to the caption.
We demonstrate our model's performance with multiple VLMs and through a range
of ablations detailing the impact of different architectural choices. Our full
model achieves a competitive CIDEr score of $136.9$ on the COCO Karpathy split.
| no_new_dataset | 0.945751 |
2503.08505 | Fan Wu | Fan Wu, Sijun Dong, Xiaoliang Meng | CFNet: Optimizing Remote Sensing Change Detection through Content-Aware
Enhancement | 17 pages, 12 figures | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Change detection is a crucial and widely applied task in remote sensing,
aimed at identifying and analyzing changes occurring in the same geographical
area over time. Due to variability in acquisition conditions, bi-temporal
remote sensing images often exhibit significant differences in image style.
Even with the powerful generalization capabilities of DNNs, these unpredictable
style variations between bi-temporal images inevitably affect model's ability
to accurately detect changed areas. To address issue above, we propose the
Content Focuser Network (CFNet), which takes content-aware strategy as a key
insight. CFNet employs EfficientNet-B5 as the backbone for feature extraction.
To enhance the model's focus on the content features of images while mitigating
the misleading effects of style features, we develop a constraint strategy that
prioritizes the content features of bi-temporal images, termed Content-Aware.
Furthermore, to enable the model to flexibly focus on changed and unchanged
areas according to the requirements of different stages, we design a
reweighting module based on the cosine distance between bi-temporal image
features, termed Focuser. CFNet achieve outstanding performance across three
well-known change detection datasets: CLCD (F1: 81.41%, IoU: 68.65%), LEVIR-CD
(F1: 92.18%, IoU: 85.49%), and SYSU-CD (F1: 82.89%, IoU: 70.78%). The code and
pretrained models of CFNet are publicly released at
https://github.com/wifiBlack/CFNet.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 14:56:11 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Wu",
"Fan",
""
],
[
"Dong",
"Sijun",
""
],
[
"Meng",
"Xiaoliang",
""
]
]
| TITLE: CFNet: Optimizing Remote Sensing Change Detection through Content-Aware
Enhancement
ABSTRACT: Change detection is a crucial and widely applied task in remote sensing,
aimed at identifying and analyzing changes occurring in the same geographical
area over time. Due to variability in acquisition conditions, bi-temporal
remote sensing images often exhibit significant differences in image style.
Even with the powerful generalization capabilities of DNNs, these unpredictable
style variations between bi-temporal images inevitably affect model's ability
to accurately detect changed areas. To address issue above, we propose the
Content Focuser Network (CFNet), which takes content-aware strategy as a key
insight. CFNet employs EfficientNet-B5 as the backbone for feature extraction.
To enhance the model's focus on the content features of images while mitigating
the misleading effects of style features, we develop a constraint strategy that
prioritizes the content features of bi-temporal images, termed Content-Aware.
Furthermore, to enable the model to flexibly focus on changed and unchanged
areas according to the requirements of different stages, we design a
reweighting module based on the cosine distance between bi-temporal image
features, termed Focuser. CFNet achieve outstanding performance across three
well-known change detection datasets: CLCD (F1: 81.41%, IoU: 68.65%), LEVIR-CD
(F1: 92.18%, IoU: 85.49%), and SYSU-CD (F1: 82.89%, IoU: 70.78%). The code and
pretrained models of CFNet are publicly released at
https://github.com/wifiBlack/CFNet.
| no_new_dataset | 0.950641 |
2503.08506 | Xian Gao | Xian Gao, Jiacheng Ruan, Jingsheng Gao, Ting Liu and Yuzhuo Fu | ReviewAgents: Bridging the Gap Between Human and AI-Generated Paper
Reviews | Work in progress | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Academic paper review is a critical yet time-consuming task within the
research community. With the increasing volume of academic publications,
automating the review process has become a significant challenge. The primary
issue lies in generating comprehensive, accurate, and reasoning-consistent
review comments that align with human reviewers' judgments. In this paper, we
address this challenge by proposing ReviewAgents, a framework that leverages
large language models (LLMs) to generate academic paper reviews. We first
introduce a novel dataset, Review-CoT, consisting of 142k review comments,
designed for training LLM agents. This dataset emulates the structured
reasoning process of human reviewers-summarizing the paper, referencing
relevant works, identifying strengths and weaknesses, and generating a review
conclusion. Building upon this, we train LLM reviewer agents capable of
structured reasoning using a relevant-paper-aware training method. Furthermore,
we construct ReviewAgents, a multi-role, multi-LLM agent review framework, to
enhance the review comment generation process. Additionally, we propose
ReviewBench, a benchmark for evaluating the review comments generated by LLMs.
Our experimental results on ReviewBench demonstrate that while existing LLMs
exhibit a certain degree of potential for automating the review process, there
remains a gap when compared to human-generated reviews. Moreover, our
ReviewAgents framework further narrows this gap, outperforming advanced LLMs in
generating review comments.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 14:56:58 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Gao",
"Xian",
""
],
[
"Ruan",
"Jiacheng",
""
],
[
"Gao",
"Jingsheng",
""
],
[
"Liu",
"Ting",
""
],
[
"Fu",
"Yuzhuo",
""
]
]
| TITLE: ReviewAgents: Bridging the Gap Between Human and AI-Generated Paper
Reviews
ABSTRACT: Academic paper review is a critical yet time-consuming task within the
research community. With the increasing volume of academic publications,
automating the review process has become a significant challenge. The primary
issue lies in generating comprehensive, accurate, and reasoning-consistent
review comments that align with human reviewers' judgments. In this paper, we
address this challenge by proposing ReviewAgents, a framework that leverages
large language models (LLMs) to generate academic paper reviews. We first
introduce a novel dataset, Review-CoT, consisting of 142k review comments,
designed for training LLM agents. This dataset emulates the structured
reasoning process of human reviewers-summarizing the paper, referencing
relevant works, identifying strengths and weaknesses, and generating a review
conclusion. Building upon this, we train LLM reviewer agents capable of
structured reasoning using a relevant-paper-aware training method. Furthermore,
we construct ReviewAgents, a multi-role, multi-LLM agent review framework, to
enhance the review comment generation process. Additionally, we propose
ReviewBench, a benchmark for evaluating the review comments generated by LLMs.
Our experimental results on ReviewBench demonstrate that while existing LLMs
exhibit a certain degree of potential for automating the review process, there
remains a gap when compared to human-generated reviews. Moreover, our
ReviewAgents framework further narrows this gap, outperforming advanced LLMs in
generating review comments.
| new_dataset | 0.962497 |
2503.08507 | Qing Jiang | Qing Jiang, Lin Wu, Zhaoyang Zeng, Tianhe Ren, Yuda Xiong, Yihao Chen,
Qin Liu, Lei Zhang | Referring to Any Person | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Humans are undoubtedly the most important participants in computer vision,
and the ability to detect any individual given a natural language description,
a task we define as referring to any person, holds substantial practical value.
However, we find that existing models generally fail to achieve real-world
usability, and current benchmarks are limited by their focus on one-to-one
referring, that hinder progress in this area. In this work, we revisit this
task from three critical perspectives: task definition, dataset design, and
model architecture. We first identify five aspects of referable entities and
three distinctive characteristics of this task. Next, we introduce HumanRef, a
novel dataset designed to tackle these challenges and better reflect real-world
applications. From a model design perspective, we integrate a multimodal large
language model with an object detection framework, constructing a robust
referring model named RexSeek. Experimental results reveal that
state-of-the-art models, which perform well on commonly used benchmarks like
RefCOCO/+/g, struggle with HumanRef due to their inability to detect multiple
individuals. In contrast, RexSeek not only excels in human referring but also
generalizes effectively to common object referring, making it broadly
applicable across various perception tasks. Code is available at
https://github.com/IDEA-Research/RexSeek
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 14:57:14 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Jiang",
"Qing",
""
],
[
"Wu",
"Lin",
""
],
[
"Zeng",
"Zhaoyang",
""
],
[
"Ren",
"Tianhe",
""
],
[
"Xiong",
"Yuda",
""
],
[
"Chen",
"Yihao",
""
],
[
"Liu",
"Qin",
""
],
[
"Zhang",
"Lei",
""
]
]
| TITLE: Referring to Any Person
ABSTRACT: Humans are undoubtedly the most important participants in computer vision,
and the ability to detect any individual given a natural language description,
a task we define as referring to any person, holds substantial practical value.
However, we find that existing models generally fail to achieve real-world
usability, and current benchmarks are limited by their focus on one-to-one
referring, that hinder progress in this area. In this work, we revisit this
task from three critical perspectives: task definition, dataset design, and
model architecture. We first identify five aspects of referable entities and
three distinctive characteristics of this task. Next, we introduce HumanRef, a
novel dataset designed to tackle these challenges and better reflect real-world
applications. From a model design perspective, we integrate a multimodal large
language model with an object detection framework, constructing a robust
referring model named RexSeek. Experimental results reveal that
state-of-the-art models, which perform well on commonly used benchmarks like
RefCOCO/+/g, struggle with HumanRef due to their inability to detect multiple
individuals. In contrast, RexSeek not only excels in human referring but also
generalizes effectively to common object referring, making it broadly
applicable across various perception tasks. Code is available at
https://github.com/IDEA-Research/RexSeek
| new_dataset | 0.960473 |
2503.08508 | Weijie Zhou | Weijie Zhou (1,2), Yi Peng (2), Manli Tao (2), Chaoyang Zhao (2,3),
Honghui Dong (1), Ming Tang (2), Jinqiao Wang (2,3) ((1) School of Traffic
and Transportation, Beijing Jiaotong University, (2) Foundation Model
Research Center, Institute of Automation, Chinese Academy of Sciences, (3)
objecteye.Inc) | LightPlanner: Unleashing the Reasoning Capabilities of Lightweight Large
Language Models in Task Planning | null | null | null | null | cs.RO | http://creativecommons.org/licenses/by/4.0/ | In recent years, lightweight large language models (LLMs) have garnered
significant attention in the robotics field due to their low computational
resource requirements and suitability for edge deployment. However, in task
planning -- particularly for complex tasks that involve dynamic semantic logic
reasoning -- lightweight LLMs have underperformed. To address this limitation,
we propose a novel task planner, LightPlanner, which enhances the performance
of lightweight LLMs in complex task planning by fully leveraging their
reasoning capabilities. Unlike conventional planners that use fixed skill
templates, LightPlanner controls robot actions via parameterized function
calls, dynamically generating parameter values. This approach allows for
fine-grained skill control and improves task planning success rates in complex
scenarios. Furthermore, we introduce hierarchical deep reasoning. Before
generating each action decision step, LightPlanner thoroughly considers three
levels: action execution (feedback verification), semantic parsing (goal
consistency verification), and parameter generation (parameter validity
verification). This ensures the correctness of subsequent action controls.
Additionally, we incorporate a memory module to store historical actions,
thereby reducing context length and enhancing planning efficiency for long-term
tasks. We train the LightPlanner-1.5B model on our LightPlan-40k dataset, which
comprises 40,000 action controls across tasks with 2 to 13 action steps.
Experiments demonstrate that our model achieves the highest task success rate
despite having the smallest number of parameters. In tasks involving spatial
semantic reasoning, the success rate exceeds that of ReAct by 14.9 percent.
Moreover, we demonstrate LightPlanner's potential to operate on edge devices.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 14:57:53 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Zhou",
"Weijie",
""
],
[
"Peng",
"Yi",
""
],
[
"Tao",
"Manli",
""
],
[
"Zhao",
"Chaoyang",
""
],
[
"Dong",
"Honghui",
""
],
[
"Tang",
"Ming",
""
],
[
"Wang",
"Jinqiao",
""
]
]
| TITLE: LightPlanner: Unleashing the Reasoning Capabilities of Lightweight Large
Language Models in Task Planning
ABSTRACT: In recent years, lightweight large language models (LLMs) have garnered
significant attention in the robotics field due to their low computational
resource requirements and suitability for edge deployment. However, in task
planning -- particularly for complex tasks that involve dynamic semantic logic
reasoning -- lightweight LLMs have underperformed. To address this limitation,
we propose a novel task planner, LightPlanner, which enhances the performance
of lightweight LLMs in complex task planning by fully leveraging their
reasoning capabilities. Unlike conventional planners that use fixed skill
templates, LightPlanner controls robot actions via parameterized function
calls, dynamically generating parameter values. This approach allows for
fine-grained skill control and improves task planning success rates in complex
scenarios. Furthermore, we introduce hierarchical deep reasoning. Before
generating each action decision step, LightPlanner thoroughly considers three
levels: action execution (feedback verification), semantic parsing (goal
consistency verification), and parameter generation (parameter validity
verification). This ensures the correctness of subsequent action controls.
Additionally, we incorporate a memory module to store historical actions,
thereby reducing context length and enhancing planning efficiency for long-term
tasks. We train the LightPlanner-1.5B model on our LightPlan-40k dataset, which
comprises 40,000 action controls across tasks with 2 to 13 action steps.
Experiments demonstrate that our model achieves the highest task success rate
despite having the smallest number of parameters. In tasks involving spatial
semantic reasoning, the success rate exceeds that of ReAct by 14.9 percent.
Moreover, we demonstrate LightPlanner's potential to operate on edge devices.
| new_dataset | 0.958343 |
2503.08510 | Da-Wei Zhou | Da-Wei Zhou, Kai-Wen Li, Jingyi Ning, Han-Jia Ye, Lijun Zhang,
De-Chuan Zhan | External Knowledge Injection for CLIP-Based Class-Incremental Learning | Code is available at: https://github.com/RenaissCode/ENGINE | null | null | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Class-Incremental Learning (CIL) enables learning systems to continuously
adapt to evolving data streams. With the advancement of pre-training,
leveraging pre-trained vision-language models (e.g., CLIP) offers a promising
starting point for CIL. However, CLIP makes decisions by matching visual
embeddings to class names, overlooking the rich contextual information conveyed
through language. For instance, the concept of ``cat'' can be decomposed into
features like tail, fur, and face for recognition. Besides, since the model is
continually updated, these detailed features are overwritten in CIL, requiring
external knowledge for compensation. In this paper, we introduce ExterNal
knowledGe INjEction (ENGINE) for CLIP-based CIL. To enhance knowledge transfer
from outside the dataset, we propose a dual-branch injection tuning framework
that encodes informative knowledge from both visual and textual modalities. The
visual branch is enhanced with data augmentation to enrich the visual features,
while the textual branch leverages GPT-4 to rewrite discriminative descriptors.
In addition to this on-the-fly knowledge injection, we also implement
post-tuning knowledge by re-ranking the prediction results during inference.
With the injected knowledge, the model can better capture informative features
for downstream tasks as data evolves. Extensive experiments demonstrate the
state-of-the-art performance of ENGINE. Code is available at:
https://github.com/RenaissCode/ENGINE
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 15:00:22 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Zhou",
"Da-Wei",
""
],
[
"Li",
"Kai-Wen",
""
],
[
"Ning",
"Jingyi",
""
],
[
"Ye",
"Han-Jia",
""
],
[
"Zhang",
"Lijun",
""
],
[
"Zhan",
"De-Chuan",
""
]
]
| TITLE: External Knowledge Injection for CLIP-Based Class-Incremental Learning
ABSTRACT: Class-Incremental Learning (CIL) enables learning systems to continuously
adapt to evolving data streams. With the advancement of pre-training,
leveraging pre-trained vision-language models (e.g., CLIP) offers a promising
starting point for CIL. However, CLIP makes decisions by matching visual
embeddings to class names, overlooking the rich contextual information conveyed
through language. For instance, the concept of ``cat'' can be decomposed into
features like tail, fur, and face for recognition. Besides, since the model is
continually updated, these detailed features are overwritten in CIL, requiring
external knowledge for compensation. In this paper, we introduce ExterNal
knowledGe INjEction (ENGINE) for CLIP-based CIL. To enhance knowledge transfer
from outside the dataset, we propose a dual-branch injection tuning framework
that encodes informative knowledge from both visual and textual modalities. The
visual branch is enhanced with data augmentation to enrich the visual features,
while the textual branch leverages GPT-4 to rewrite discriminative descriptors.
In addition to this on-the-fly knowledge injection, we also implement
post-tuning knowledge by re-ranking the prediction results during inference.
With the injected knowledge, the model can better capture informative features
for downstream tasks as data evolves. Extensive experiments demonstrate the
state-of-the-art performance of ENGINE. Code is available at:
https://github.com/RenaissCode/ENGINE
| no_new_dataset | 0.945551 |
2503.08512 | Zhuoyuan Li | Zhuoyuan Li, Jiahao Lu, Jiacheng Deng, Hanzhi Chang, Lifan Wu, Yanzhe
Liang, Tianzhu Zhang | SAS: Segment Any 3D Scene with Integrated 2D Priors | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | The open vocabulary capability of 3D models is increasingly valued, as
traditional methods with models trained with fixed categories fail to recognize
unseen objects in complex dynamic 3D scenes. In this paper, we propose a simple
yet effective approach, SAS, to integrate the open vocabulary capability of
multiple 2D models and migrate it to 3D domain. Specifically, we first propose
Model Alignment via Text to map different 2D models into the same embedding
space using text as a bridge. Then, we propose Annotation-Free Model Capability
Construction to explicitly quantify the 2D model's capability of recognizing
different categories using diffusion models. Following this, point cloud
features from different 2D models are fused with the guide of constructed model
capabilities. Finally, the integrated 2D open vocabulary capability is
transferred to 3D domain through feature distillation. SAS outperforms previous
methods by a large margin across multiple datasets, including ScanNet v2,
Matterport3D, and nuScenes, while its generalizability is further validated on
downstream tasks, e.g., gaussian segmentation and instance segmentation.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 15:01:54 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Li",
"Zhuoyuan",
""
],
[
"Lu",
"Jiahao",
""
],
[
"Deng",
"Jiacheng",
""
],
[
"Chang",
"Hanzhi",
""
],
[
"Wu",
"Lifan",
""
],
[
"Liang",
"Yanzhe",
""
],
[
"Zhang",
"Tianzhu",
""
]
]
| TITLE: SAS: Segment Any 3D Scene with Integrated 2D Priors
ABSTRACT: The open vocabulary capability of 3D models is increasingly valued, as
traditional methods with models trained with fixed categories fail to recognize
unseen objects in complex dynamic 3D scenes. In this paper, we propose a simple
yet effective approach, SAS, to integrate the open vocabulary capability of
multiple 2D models and migrate it to 3D domain. Specifically, we first propose
Model Alignment via Text to map different 2D models into the same embedding
space using text as a bridge. Then, we propose Annotation-Free Model Capability
Construction to explicitly quantify the 2D model's capability of recognizing
different categories using diffusion models. Following this, point cloud
features from different 2D models are fused with the guide of constructed model
capabilities. Finally, the integrated 2D open vocabulary capability is
transferred to 3D domain through feature distillation. SAS outperforms previous
methods by a large margin across multiple datasets, including ScanNet v2,
Matterport3D, and nuScenes, while its generalizability is further validated on
downstream tasks, e.g., gaussian segmentation and instance segmentation.
| no_new_dataset | 0.9463 |
2503.08515 | David Vallmanya Poch | David Vallmanya Poch, Yorick Estievenart, Elnura Zhalieva, Sukanya
Patra, Mohammad Yaqub, Souhaib Ben Taieb | Segmentation-Guided CT Synthesis with Pixel-Wise Conformal Uncertainty
Bounds | MICCAI 2025 Conference Submission. Follows the required LNCS format.
12 pages including references. Contains 4 figures and 1 table | null | null | null | cs.CV physics.med-ph | http://creativecommons.org/licenses/by/4.0/ | Accurate dose calculations in proton therapy rely on high-quality CT images.
While planning CTs (pCTs) serve as a reference for dosimetric planning, Cone
Beam CT (CBCT) is used throughout Adaptive Radiotherapy (ART) to generate sCTs
for improved dose calculations. Despite its lower cost and reduced radiation
exposure advantages, CBCT suffers from severe artefacts and poor image quality,
making it unsuitable for precise dosimetry. Deep learning-based CBCT-to-CT
translation has emerged as a promising approach. Still, existing methods often
introduce anatomical inconsistencies and lack reliable uncertainty estimates,
limiting their clinical adoption. To bridge this gap, we propose STF-RUE, a
novel framework integrating two key components. First, STF, a
segmentation-guided CBCT-to-CT translation method that enhances anatomical
consistency by leveraging segmentation priors extracted from pCTs. Second, RUE,
a conformal prediction method that augments predicted CTs with pixel-wise
conformal prediction intervals, providing clinicians with robust reliability
indicator. Comprehensive experiments using UNet++ and Fast-DDPM on two
benchmark datasets demonstrate that STF-RUE significantly improves translation
accuracy, as measured by a novel soft-tissue-focused metric designed for
precise dose computation. Additionally, STF-RUE provides better-calibrated
uncertainty sets for synthetic CT, reinforcing trust in synthetic CTs. By
addressing both anatomical fidelity and uncertainty quantification, STF-RUE
marks a crucial step toward safer and more effective adaptive proton therapy.
Code is available at
https://anonymous.4open.science/r/cbct2ct_translation-B2D9/.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 15:07:16 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Poch",
"David Vallmanya",
""
],
[
"Estievenart",
"Yorick",
""
],
[
"Zhalieva",
"Elnura",
""
],
[
"Patra",
"Sukanya",
""
],
[
"Yaqub",
"Mohammad",
""
],
[
"Taieb",
"Souhaib Ben",
""
]
]
| TITLE: Segmentation-Guided CT Synthesis with Pixel-Wise Conformal Uncertainty
Bounds
ABSTRACT: Accurate dose calculations in proton therapy rely on high-quality CT images.
While planning CTs (pCTs) serve as a reference for dosimetric planning, Cone
Beam CT (CBCT) is used throughout Adaptive Radiotherapy (ART) to generate sCTs
for improved dose calculations. Despite its lower cost and reduced radiation
exposure advantages, CBCT suffers from severe artefacts and poor image quality,
making it unsuitable for precise dosimetry. Deep learning-based CBCT-to-CT
translation has emerged as a promising approach. Still, existing methods often
introduce anatomical inconsistencies and lack reliable uncertainty estimates,
limiting their clinical adoption. To bridge this gap, we propose STF-RUE, a
novel framework integrating two key components. First, STF, a
segmentation-guided CBCT-to-CT translation method that enhances anatomical
consistency by leveraging segmentation priors extracted from pCTs. Second, RUE,
a conformal prediction method that augments predicted CTs with pixel-wise
conformal prediction intervals, providing clinicians with robust reliability
indicator. Comprehensive experiments using UNet++ and Fast-DDPM on two
benchmark datasets demonstrate that STF-RUE significantly improves translation
accuracy, as measured by a novel soft-tissue-focused metric designed for
precise dose computation. Additionally, STF-RUE provides better-calibrated
uncertainty sets for synthetic CT, reinforcing trust in synthetic CTs. By
addressing both anatomical fidelity and uncertainty quantification, STF-RUE
marks a crucial step toward safer and more effective adaptive proton therapy.
Code is available at
https://anonymous.4open.science/r/cbct2ct_translation-B2D9/.
| no_new_dataset | 0.950732 |
2503.08529 | Ryan Wong | Ryan Wong, Necati Cihan Camgoz, Richard Bowden | SignRep: Enhancing Self-Supervised Sign Representations | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Sign language representation learning presents unique challenges due to the
complex spatio-temporal nature of signs and the scarcity of labeled datasets.
Existing methods often rely either on models pre-trained on general visual
tasks, that lack sign-specific features, or use complex multimodal and
multi-branch architectures. To bridge this gap, we introduce a scalable,
self-supervised framework for sign representation learning. We leverage
important inductive (sign) priors during the training of our RGB model. To do
this, we leverage simple but important cues based on skeletons while
pretraining a masked autoencoder. These sign specific priors alongside feature
regularization and an adversarial style agnostic loss provide a powerful
backbone. Notably, our model does not require skeletal keypoints during
inference, avoiding the limitations of keypoint-based models during downstream
tasks. When finetuned, we achieve state-of-the-art performance for sign
recognition on the WLASL, ASL-Citizen and NMFs-CSL datasets, using a simpler
architecture and with only a single-modality. Beyond recognition, our frozen
model excels in sign dictionary retrieval and sign translation, surpassing
standard MAE pretraining and skeletal-based representations in retrieval. It
also reduces computational costs for training existing sign translation models
while maintaining strong performance on Phoenix2014T, CSL-Daily and How2Sign.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 15:20:01 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Wong",
"Ryan",
""
],
[
"Camgoz",
"Necati Cihan",
""
],
[
"Bowden",
"Richard",
""
]
]
| TITLE: SignRep: Enhancing Self-Supervised Sign Representations
ABSTRACT: Sign language representation learning presents unique challenges due to the
complex spatio-temporal nature of signs and the scarcity of labeled datasets.
Existing methods often rely either on models pre-trained on general visual
tasks, that lack sign-specific features, or use complex multimodal and
multi-branch architectures. To bridge this gap, we introduce a scalable,
self-supervised framework for sign representation learning. We leverage
important inductive (sign) priors during the training of our RGB model. To do
this, we leverage simple but important cues based on skeletons while
pretraining a masked autoencoder. These sign specific priors alongside feature
regularization and an adversarial style agnostic loss provide a powerful
backbone. Notably, our model does not require skeletal keypoints during
inference, avoiding the limitations of keypoint-based models during downstream
tasks. When finetuned, we achieve state-of-the-art performance for sign
recognition on the WLASL, ASL-Citizen and NMFs-CSL datasets, using a simpler
architecture and with only a single-modality. Beyond recognition, our frozen
model excels in sign dictionary retrieval and sign translation, surpassing
standard MAE pretraining and skeletal-based representations in retrieval. It
also reduces computational costs for training existing sign translation models
while maintaining strong performance on Phoenix2014T, CSL-Daily and How2Sign.
| no_new_dataset | 0.946498 |
2503.08532 | Julian Aron Prenner | Julian Aron Prenner and Romain Robbes | Bogus Bugs, Duplicates, and Revealing Comments: Data Quality Issues in
NPR | null | null | null | null | cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The performance of a machine learning system is not only determined by the
model but also, to a substantial degree, by the data it is trained on. With the
increasing use of machine learning, issues related to data quality have become
a concern also in automated program repair research. In this position paper, we
report some of the data-related issues we have come across when working with
several large APR datasets and benchmarks, including, for instance, duplicates
or "bogus bugs". We briefly discuss the potential impact of these problems on
repair performance and propose possible remedies. We believe that more
data-focused approaches could improve the performance and robustness of current
and future APR systems.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 15:23:13 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Prenner",
"Julian Aron",
""
],
[
"Robbes",
"Romain",
""
]
]
| TITLE: Bogus Bugs, Duplicates, and Revealing Comments: Data Quality Issues in
NPR
ABSTRACT: The performance of a machine learning system is not only determined by the
model but also, to a substantial degree, by the data it is trained on. With the
increasing use of machine learning, issues related to data quality have become
a concern also in automated program repair research. In this position paper, we
report some of the data-related issues we have come across when working with
several large APR datasets and benchmarks, including, for instance, duplicates
or "bogus bugs". We briefly discuss the potential impact of these problems on
repair performance and propose possible remedies. We believe that more
data-focused approaches could improve the performance and robustness of current
and future APR systems.
| no_new_dataset | 0.956104 |
2503.08533 | Siddhant Arora | Siddhant Arora, Yifan Peng, Jiatong Shi, Jinchuan Tian, William Chen,
Shikhar Bharadwaj, Hayato Futami, Yosuke Kashiwagi, Emiru Tsunoo, Shuichiro
Shimizu, Vaibhav Srivastav, Shinji Watanabe | ESPnet-SDS: Unified Toolkit and Demo for Spoken Dialogue Systems | Accepted at NAACL 2025 Demo Track | null | null | null | cs.CL cs.SD eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Advancements in audio foundation models (FMs) have fueled interest in
end-to-end (E2E) spoken dialogue systems, but different web interfaces for each
system makes it challenging to compare and contrast them effectively. Motivated
by this, we introduce an open-source, user-friendly toolkit designed to build
unified web interfaces for various cascaded and E2E spoken dialogue systems.
Our demo further provides users with the option to get on-the-fly automated
evaluation metrics such as (1) latency, (2) ability to understand user input,
(3) coherence, diversity, and relevance of system response, and (4)
intelligibility and audio quality of system output. Using the evaluation
metrics, we compare various cascaded and E2E spoken dialogue systems with a
human-human conversation dataset as a proxy. Our analysis demonstrates that the
toolkit allows researchers to effortlessly compare and contrast different
technologies, providing valuable insights such as current E2E systems having
poorer audio quality and less diverse responses. An example demo produced using
our toolkit is publicly available here:
https://huggingface.co/spaces/Siddhant/Voice_Assistant_Demo.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 15:24:02 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Arora",
"Siddhant",
""
],
[
"Peng",
"Yifan",
""
],
[
"Shi",
"Jiatong",
""
],
[
"Tian",
"Jinchuan",
""
],
[
"Chen",
"William",
""
],
[
"Bharadwaj",
"Shikhar",
""
],
[
"Futami",
"Hayato",
""
],
[
"Kashiwagi",
"Yosuke",
""
],
[
"Tsunoo",
"Emiru",
""
],
[
"Shimizu",
"Shuichiro",
""
],
[
"Srivastav",
"Vaibhav",
""
],
[
"Watanabe",
"Shinji",
""
]
]
| TITLE: ESPnet-SDS: Unified Toolkit and Demo for Spoken Dialogue Systems
ABSTRACT: Advancements in audio foundation models (FMs) have fueled interest in
end-to-end (E2E) spoken dialogue systems, but different web interfaces for each
system makes it challenging to compare and contrast them effectively. Motivated
by this, we introduce an open-source, user-friendly toolkit designed to build
unified web interfaces for various cascaded and E2E spoken dialogue systems.
Our demo further provides users with the option to get on-the-fly automated
evaluation metrics such as (1) latency, (2) ability to understand user input,
(3) coherence, diversity, and relevance of system response, and (4)
intelligibility and audio quality of system output. Using the evaluation
metrics, we compare various cascaded and E2E spoken dialogue systems with a
human-human conversation dataset as a proxy. Our analysis demonstrates that the
toolkit allows researchers to effortlessly compare and contrast different
technologies, providing valuable insights such as current E2E systems having
poorer audio quality and less diverse responses. An example demo produced using
our toolkit is publicly available here:
https://huggingface.co/spaces/Siddhant/Voice_Assistant_Demo.
| no_new_dataset | 0.945901 |
2503.08534 | Mingshi Li | Mingshi Li, Dusan Grujicic, Ben Somers, Stien Heremans, Steven De
Saeger, Matthew B. Blaschko | ChromaFormer: A Scalable and Accurate Transformer Architecture for Land
Cover Classification | null | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Remote sensing imagery from systems such as Sentinel provides full coverage
of the Earth's surface at around 10-meter resolution. The remote sensing
community has transitioned to extensive use of deep learning models due to
their high performance on benchmarks such as the UCMerced and ISPRS Vaihingen
datasets. Convolutional models such as UNet and ResNet variations are commonly
employed for remote sensing but typically only accept three channels, as they
were developed for RGB imagery, while satellite systems provide more than ten.
Recently, several transformer architectures have been proposed for remote
sensing, but they have not been extensively benchmarked and are typically used
on small datasets such as Salinas Valley. Meanwhile, it is becoming feasible to
obtain dense spatial land-use labels for entire first-level administrative
divisions of some countries. Scaling law observations suggest that
substantially larger multi-spectral transformer models could provide a
significant leap in remote sensing performance in these settings.
In this work, we propose ChromaFormer, a family of multi-spectral transformer
models, which we evaluate across orders of magnitude differences in model
parameters to assess their performance and scaling effectiveness on a densely
labeled imagery dataset of Flanders, Belgium, covering more than 13,500 km^2
and containing 15 classes. We propose a novel multi-spectral attention strategy
and demonstrate its effectiveness through ablations. Furthermore, we show that
models many orders of magnitude larger than conventional architectures, such as
UNet, lead to substantial accuracy improvements: a UNet++ model with 23M
parameters achieves less than 65% accuracy, while a multi-spectral transformer
with 655M parameters achieves over 95% accuracy on the Biological Valuation Map
of Flanders.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 15:24:50 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Li",
"Mingshi",
""
],
[
"Grujicic",
"Dusan",
""
],
[
"Somers",
"Ben",
""
],
[
"Heremans",
"Stien",
""
],
[
"De Saeger",
"Steven",
""
],
[
"Blaschko",
"Matthew B.",
""
]
]
| TITLE: ChromaFormer: A Scalable and Accurate Transformer Architecture for Land
Cover Classification
ABSTRACT: Remote sensing imagery from systems such as Sentinel provides full coverage
of the Earth's surface at around 10-meter resolution. The remote sensing
community has transitioned to extensive use of deep learning models due to
their high performance on benchmarks such as the UCMerced and ISPRS Vaihingen
datasets. Convolutional models such as UNet and ResNet variations are commonly
employed for remote sensing but typically only accept three channels, as they
were developed for RGB imagery, while satellite systems provide more than ten.
Recently, several transformer architectures have been proposed for remote
sensing, but they have not been extensively benchmarked and are typically used
on small datasets such as Salinas Valley. Meanwhile, it is becoming feasible to
obtain dense spatial land-use labels for entire first-level administrative
divisions of some countries. Scaling law observations suggest that
substantially larger multi-spectral transformer models could provide a
significant leap in remote sensing performance in these settings.
In this work, we propose ChromaFormer, a family of multi-spectral transformer
models, which we evaluate across orders of magnitude differences in model
parameters to assess their performance and scaling effectiveness on a densely
labeled imagery dataset of Flanders, Belgium, covering more than 13,500 km^2
and containing 15 classes. We propose a novel multi-spectral attention strategy
and demonstrate its effectiveness through ablations. Furthermore, we show that
models many orders of magnitude larger than conventional architectures, such as
UNet, lead to substantial accuracy improvements: a UNet++ model with 23M
parameters achieves less than 65% accuracy, while a multi-spectral transformer
with 655M parameters achieves over 95% accuracy on the Biological Valuation Map
of Flanders.
| no_new_dataset | 0.950041 |
2503.08540 | Soham Deshmukh | Soham Deshmukh, Satvik Dixit, Rita Singh, Bhiksha Raj | Mellow: a small audio language model for reasoning | Checkpoint and dataset available at:
https://github.com/soham97/mellow | null | null | null | cs.SD cs.AI eess.AS | http://creativecommons.org/licenses/by/4.0/ | Multimodal Audio-Language Models (ALMs) can understand and reason over both
audio and text. Typically, reasoning performance correlates with model size,
with the best results achieved by models exceeding 8 billion parameters.
However, no prior work has explored enabling small audio-language models to
perform reasoning tasks, despite the potential applications for edge devices.
To address this gap, we introduce Mellow, a small Audio-Language Model
specifically designed for reasoning. Mellow achieves state-of-the-art
performance among existing small audio-language models and surpasses several
larger models in reasoning capabilities. For instance, Mellow scores 52.11 on
MMAU, comparable to SoTA Qwen2 Audio (which scores 52.5) while using 50 times
fewer parameters and being trained on 60 times less data (audio hrs). To train
Mellow, we introduce ReasonAQA, a dataset designed to enhance audio-grounded
reasoning in models. It consists of a mixture of existing datasets (30% of the
data) and synthetically generated data (70%). The synthetic dataset is derived
from audio captioning datasets, where Large Language Models (LLMs) generate
detailed and multiple-choice questions focusing on audio events, objects,
acoustic scenes, signal properties, semantics, and listener emotions. To
evaluate Mellow's reasoning ability, we benchmark it on a diverse set of tasks,
assessing on both in-distribution and out-of-distribution data, including audio
understanding, deductive reasoning, and comparative reasoning. Finally, we
conduct extensive ablation studies to explore the impact of projection layer
choices, synthetic data generation methods, and language model pretraining on
reasoning performance. Our training dataset, findings, and baseline pave the
way for developing small ALMs capable of reasoning.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 15:29:00 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Deshmukh",
"Soham",
""
],
[
"Dixit",
"Satvik",
""
],
[
"Singh",
"Rita",
""
],
[
"Raj",
"Bhiksha",
""
]
]
| TITLE: Mellow: a small audio language model for reasoning
ABSTRACT: Multimodal Audio-Language Models (ALMs) can understand and reason over both
audio and text. Typically, reasoning performance correlates with model size,
with the best results achieved by models exceeding 8 billion parameters.
However, no prior work has explored enabling small audio-language models to
perform reasoning tasks, despite the potential applications for edge devices.
To address this gap, we introduce Mellow, a small Audio-Language Model
specifically designed for reasoning. Mellow achieves state-of-the-art
performance among existing small audio-language models and surpasses several
larger models in reasoning capabilities. For instance, Mellow scores 52.11 on
MMAU, comparable to SoTA Qwen2 Audio (which scores 52.5) while using 50 times
fewer parameters and being trained on 60 times less data (audio hrs). To train
Mellow, we introduce ReasonAQA, a dataset designed to enhance audio-grounded
reasoning in models. It consists of a mixture of existing datasets (30% of the
data) and synthetically generated data (70%). The synthetic dataset is derived
from audio captioning datasets, where Large Language Models (LLMs) generate
detailed and multiple-choice questions focusing on audio events, objects,
acoustic scenes, signal properties, semantics, and listener emotions. To
evaluate Mellow's reasoning ability, we benchmark it on a diverse set of tasks,
assessing on both in-distribution and out-of-distribution data, including audio
understanding, deductive reasoning, and comparative reasoning. Finally, we
conduct extensive ablation studies to explore the impact of projection layer
choices, synthetic data generation methods, and language model pretraining on
reasoning performance. Our training dataset, findings, and baseline pave the
way for developing small ALMs capable of reasoning.
| new_dataset | 0.968827 |
2503.08548 | Peng Hao | Peng Hao, Chaofan Zhang, Dingzhe Li, Xiaoge Cao, Xiaoshuai Hao,
Shaowei Cui, Shuo Wang | TLA: Tactile-Language-Action Model for Contact-Rich Manipulation | null | null | null | null | cs.RO cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Significant progress has been made in vision-language models. However,
language-conditioned robotic manipulation for contact-rich tasks remains
underexplored, particularly in terms of tactile sensing. To address this gap,
we introduce the Tactile-Language-Action (TLA) model, which effectively
processes sequential tactile feedback via cross-modal language grounding to
enable robust policy generation in contact-intensive scenarios. In addition, we
construct a comprehensive dataset that contains 24k pairs of tactile action
instruction data, customized for fingertip peg-in-hole assembly, providing
essential resources for TLA training and evaluation. Our results show that TLA
significantly outperforms traditional imitation learning methods (e.g.,
diffusion policy) in terms of effective action generation and action accuracy,
while demonstrating strong generalization capabilities by achieving over 85\%
success rate on previously unseen assembly clearances and peg shapes. We
publicly release all data and code in the hope of advancing research in
language-conditioned tactile manipulation skill learning. Project website:
https://sites.google.com/view/tactile-language-action/
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 15:36:28 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Hao",
"Peng",
""
],
[
"Zhang",
"Chaofan",
""
],
[
"Li",
"Dingzhe",
""
],
[
"Cao",
"Xiaoge",
""
],
[
"Hao",
"Xiaoshuai",
""
],
[
"Cui",
"Shaowei",
""
],
[
"Wang",
"Shuo",
""
]
]
| TITLE: TLA: Tactile-Language-Action Model for Contact-Rich Manipulation
ABSTRACT: Significant progress has been made in vision-language models. However,
language-conditioned robotic manipulation for contact-rich tasks remains
underexplored, particularly in terms of tactile sensing. To address this gap,
we introduce the Tactile-Language-Action (TLA) model, which effectively
processes sequential tactile feedback via cross-modal language grounding to
enable robust policy generation in contact-intensive scenarios. In addition, we
construct a comprehensive dataset that contains 24k pairs of tactile action
instruction data, customized for fingertip peg-in-hole assembly, providing
essential resources for TLA training and evaluation. Our results show that TLA
significantly outperforms traditional imitation learning methods (e.g.,
diffusion policy) in terms of effective action generation and action accuracy,
while demonstrating strong generalization capabilities by achieving over 85\%
success rate on previously unseen assembly clearances and peg shapes. We
publicly release all data and code in the hope of advancing research in
language-conditioned tactile manipulation skill learning. Project website:
https://sites.google.com/view/tactile-language-action/
| new_dataset | 0.960212 |
2503.08551 | Wanyong Feng | Wanyong Feng, Peter Tran, Stephen Sireci, Andrew Lan | Reasoning and Sampling-Augmented MCQ Difficulty Prediction via LLMs | null | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | The difficulty of multiple-choice questions (MCQs) is a crucial factor for
educational assessments. Predicting MCQ difficulty is challenging since it
requires understanding both the complexity of reaching the correct option and
the plausibility of distractors, i.e., incorrect options. In this paper, we
propose a novel, two-stage method to predict the difficulty of MCQs. First, to
better estimate the complexity of each MCQ, we use large language models (LLMs)
to augment the reasoning steps required to reach each option. We use not just
the MCQ itself but also these reasoning steps as input to predict the
difficulty. Second, to capture the plausibility of distractors, we sample
knowledge levels from a distribution to account for variation among students
responding to the MCQ. This setup, inspired by item response theory (IRT),
enable us to estimate the likelihood of students selecting each (both correct
and incorrect) option. We align these predictions with their ground truth
values, using a Kullback-Leibler (KL) divergence-based regularization
objective, and use estimated likelihoods to predict MCQ difficulty. We evaluate
our method on two real-world \emph{math} MCQ and response datasets with ground
truth difficulty values estimated using IRT. Experimental results show that our
method outperforms all baselines, up to a 28.3\% reduction in mean squared
error and a 34.6\% improvement in the coefficient of determination. We also
qualitatively discuss how our novel method results in higher accuracy in
predicting MCQ difficulty.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 15:39:43 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Feng",
"Wanyong",
""
],
[
"Tran",
"Peter",
""
],
[
"Sireci",
"Stephen",
""
],
[
"Lan",
"Andrew",
""
]
]
| TITLE: Reasoning and Sampling-Augmented MCQ Difficulty Prediction via LLMs
ABSTRACT: The difficulty of multiple-choice questions (MCQs) is a crucial factor for
educational assessments. Predicting MCQ difficulty is challenging since it
requires understanding both the complexity of reaching the correct option and
the plausibility of distractors, i.e., incorrect options. In this paper, we
propose a novel, two-stage method to predict the difficulty of MCQs. First, to
better estimate the complexity of each MCQ, we use large language models (LLMs)
to augment the reasoning steps required to reach each option. We use not just
the MCQ itself but also these reasoning steps as input to predict the
difficulty. Second, to capture the plausibility of distractors, we sample
knowledge levels from a distribution to account for variation among students
responding to the MCQ. This setup, inspired by item response theory (IRT),
enable us to estimate the likelihood of students selecting each (both correct
and incorrect) option. We align these predictions with their ground truth
values, using a Kullback-Leibler (KL) divergence-based regularization
objective, and use estimated likelihoods to predict MCQ difficulty. We evaluate
our method on two real-world \emph{math} MCQ and response datasets with ground
truth difficulty values estimated using IRT. Experimental results show that our
method outperforms all baselines, up to a 28.3\% reduction in mean squared
error and a 34.6\% improvement in the coefficient of determination. We also
qualitatively discuss how our novel method results in higher accuracy in
predicting MCQ difficulty.
| no_new_dataset | 0.948822 |
2503.08569 | Yixuan Weng | Minjun Zhu, Yixuan Weng, Linyi Yang, Yue Zhang | DeepReview: Improving LLM-based Paper Review with Human-like Deep
Thinking Process | null | null | null | null | cs.CL cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large Language Models (LLMs) are increasingly utilized in scientific research
assessment, particularly in automated paper review. However, existing LLM-based
review systems face significant challenges, including limited domain expertise,
hallucinated reasoning, and a lack of structured evaluation. To address these
limitations, we introduce DeepReview, a multi-stage framework designed to
emulate expert reviewers by incorporating structured analysis, literature
retrieval, and evidence-based argumentation. Using DeepReview-13K, a curated
dataset with structured annotations, we train DeepReviewer-14B, which
outperforms CycleReviewer-70B with fewer tokens. In its best mode,
DeepReviewer-14B achieves win rates of 88.21\% and 80.20\% against GPT-o1 and
DeepSeek-R1 in evaluations. Our work sets a new benchmark for LLM-based paper
review, with all resources publicly available. The code, model, dataset and
demo have be released in http://ai-researcher.net.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 15:59:43 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Zhu",
"Minjun",
""
],
[
"Weng",
"Yixuan",
""
],
[
"Yang",
"Linyi",
""
],
[
"Zhang",
"Yue",
""
]
]
| TITLE: DeepReview: Improving LLM-based Paper Review with Human-like Deep
Thinking Process
ABSTRACT: Large Language Models (LLMs) are increasingly utilized in scientific research
assessment, particularly in automated paper review. However, existing LLM-based
review systems face significant challenges, including limited domain expertise,
hallucinated reasoning, and a lack of structured evaluation. To address these
limitations, we introduce DeepReview, a multi-stage framework designed to
emulate expert reviewers by incorporating structured analysis, literature
retrieval, and evidence-based argumentation. Using DeepReview-13K, a curated
dataset with structured annotations, we train DeepReviewer-14B, which
outperforms CycleReviewer-70B with fewer tokens. In its best mode,
DeepReviewer-14B achieves win rates of 88.21\% and 80.20\% against GPT-o1 and
DeepSeek-R1 in evaluations. Our work sets a new benchmark for LLM-based paper
review, with all resources publicly available. The code, model, dataset and
demo have be released in http://ai-researcher.net.
| new_dataset | 0.951414 |
2503.08576 | Xichen Tan | Xichen Tan, Yunfan Ye, Yuanjing Luo, Qian Wan, Fang Liu, Zhiping Cai | RAG-Adapter: A Plug-and-Play RAG-enhanced Framework for Long Video
Understanding | 37 pages, 36 figures | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multi-modal Large Language Models (MLLMs) capable of video understanding are
advancing rapidly. To effectively assess their video comprehension
capabilities, long video understanding benchmarks, such as Video-MME and MLVU,
are proposed. However, these benchmarks directly use uniform frame sampling for
testing, which results in significant information loss and affects the accuracy
of the evaluations in reflecting the true abilities of MLLMs. To address this,
we propose RAG-Adapter, a plug-and-play framework that reduces information loss
during testing by sampling frames most relevant to the given question.
Additionally, we introduce a Grouped-supervised Contrastive Learning (GCL)
method to further enhance sampling effectiveness of RAG-Adapter through
fine-tuning on our constructed MMAT dataset. Finally, we test numerous baseline
MLLMs on various video understanding benchmarks, finding that RAG-Adapter
sampling consistently outperforms uniform sampling (e.g., Accuracy of GPT-4o
increases by 9.3 percent on Video-MME), providing a more accurate testing
method for long video benchmarks.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 16:10:43 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Tan",
"Xichen",
""
],
[
"Ye",
"Yunfan",
""
],
[
"Luo",
"Yuanjing",
""
],
[
"Wan",
"Qian",
""
],
[
"Liu",
"Fang",
""
],
[
"Cai",
"Zhiping",
""
]
]
| TITLE: RAG-Adapter: A Plug-and-Play RAG-enhanced Framework for Long Video
Understanding
ABSTRACT: Multi-modal Large Language Models (MLLMs) capable of video understanding are
advancing rapidly. To effectively assess their video comprehension
capabilities, long video understanding benchmarks, such as Video-MME and MLVU,
are proposed. However, these benchmarks directly use uniform frame sampling for
testing, which results in significant information loss and affects the accuracy
of the evaluations in reflecting the true abilities of MLLMs. To address this,
we propose RAG-Adapter, a plug-and-play framework that reduces information loss
during testing by sampling frames most relevant to the given question.
Additionally, we introduce a Grouped-supervised Contrastive Learning (GCL)
method to further enhance sampling effectiveness of RAG-Adapter through
fine-tuning on our constructed MMAT dataset. Finally, we test numerous baseline
MLLMs on various video understanding benchmarks, finding that RAG-Adapter
sampling consistently outperforms uniform sampling (e.g., Accuracy of GPT-4o
increases by 9.3 percent on Video-MME), providing a more accurate testing
method for long video benchmarks.
| new_dataset | 0.946101 |
2503.08585 | Shehreen Azad | Shehreen Azad, Vibhav Vineet, Yogesh Singh Rawat | HierarQ: Task-Aware Hierarchical Q-Former for Enhanced Video
Understanding | Accepted in CVPR 2025 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Despite advancements in multimodal large language models (MLLMs), current
approaches struggle in medium-to-long video understanding due to frame and
context length limitations. As a result, these models often depend on frame
sampling, which risks missing key information over time and lacks task-specific
relevance. To address these challenges, we introduce HierarQ, a task-aware
hierarchical Q-Former based framework that sequentially processes frames to
bypass the need for frame sampling, while avoiding LLM's context length
limitations. We introduce a lightweight two-stream language-guided feature
modulator to incorporate task awareness in video understanding, with the entity
stream capturing frame-level object information within a short context and the
scene stream identifying their broader interactions over longer period of time.
Each stream is supported by dedicated memory banks which enables our proposed
Hierachical Querying transformer (HierarQ) to effectively capture short and
long-term context. Extensive evaluations on 10 video benchmarks across video
understanding, question answering, and captioning tasks demonstrate HierarQ's
state-of-the-art performance across most datasets, proving its robustness and
efficiency for comprehensive video analysis.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 16:21:23 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Azad",
"Shehreen",
""
],
[
"Vineet",
"Vibhav",
""
],
[
"Rawat",
"Yogesh Singh",
""
]
]
| TITLE: HierarQ: Task-Aware Hierarchical Q-Former for Enhanced Video
Understanding
ABSTRACT: Despite advancements in multimodal large language models (MLLMs), current
approaches struggle in medium-to-long video understanding due to frame and
context length limitations. As a result, these models often depend on frame
sampling, which risks missing key information over time and lacks task-specific
relevance. To address these challenges, we introduce HierarQ, a task-aware
hierarchical Q-Former based framework that sequentially processes frames to
bypass the need for frame sampling, while avoiding LLM's context length
limitations. We introduce a lightweight two-stream language-guided feature
modulator to incorporate task awareness in video understanding, with the entity
stream capturing frame-level object information within a short context and the
scene stream identifying their broader interactions over longer period of time.
Each stream is supported by dedicated memory banks which enables our proposed
Hierachical Querying transformer (HierarQ) to effectively capture short and
long-term context. Extensive evaluations on 10 video benchmarks across video
understanding, question answering, and captioning tasks demonstrate HierarQ's
state-of-the-art performance across most datasets, proving its robustness and
efficiency for comprehensive video analysis.
| no_new_dataset | 0.943243 |
2503.08589 | Paul Calle | Paul Calle, Averi Bates, Justin C. Reynolds, Yunlong Liu, Haoyang Cui,
Sinaro Ly, Chen Wang, Qinghao Zhang, Alberto J. de Armendi, Shashank S.
Shettar, Kar Ming Fung, Qinggong Tang, Chongle Pan | Integration of nested cross-validation, automated hyperparameter
optimization, high-performance computing to reduce and quantify the variance
of test performance estimation of deep learning models | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | The variability and biases in the real-world performance benchmarking of deep
learning models for medical imaging compromise their trustworthiness for
real-world deployment. The common approach of holding out a single fixed test
set fails to quantify the variance in the estimation of test performance
metrics. This study introduces NACHOS (Nested and Automated Cross-validation
and Hyperparameter Optimization using Supercomputing) to reduce and quantify
the variance of test performance metrics of deep learning models. NACHOS
integrates Nested Cross-Validation (NCV) and Automated Hyperparameter
Optimization (AHPO) within a parallelized high-performance computing (HPC)
framework. NACHOS was demonstrated on a chest X-ray repository and an Optical
Coherence Tomography (OCT) dataset under multiple data partitioning schemes.
Beyond performance estimation, DACHOS (Deployment with Automated
Cross-validation and Hyperparameter Optimization using Supercomputing) is
introduced to leverage AHPO and cross-validation to build the final model on
the full dataset, improving expected deployment performance. The findings
underscore the importance of NCV in quantifying and reducing estimation
variance, AHPO in optimizing hyperparameters consistently across test folds,
and HPC in ensuring computational feasibility. By integrating these
methodologies, NACHOS and DACHOS provide a scalable, reproducible, and
trustworthy framework for DL model evaluation and deployment in medical
imaging.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 16:25:44 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Calle",
"Paul",
""
],
[
"Bates",
"Averi",
""
],
[
"Reynolds",
"Justin C.",
""
],
[
"Liu",
"Yunlong",
""
],
[
"Cui",
"Haoyang",
""
],
[
"Ly",
"Sinaro",
""
],
[
"Wang",
"Chen",
""
],
[
"Zhang",
"Qinghao",
""
],
[
"de Armendi",
"Alberto J.",
""
],
[
"Shettar",
"Shashank S.",
""
],
[
"Fung",
"Kar Ming",
""
],
[
"Tang",
"Qinggong",
""
],
[
"Pan",
"Chongle",
""
]
]
| TITLE: Integration of nested cross-validation, automated hyperparameter
optimization, high-performance computing to reduce and quantify the variance
of test performance estimation of deep learning models
ABSTRACT: The variability and biases in the real-world performance benchmarking of deep
learning models for medical imaging compromise their trustworthiness for
real-world deployment. The common approach of holding out a single fixed test
set fails to quantify the variance in the estimation of test performance
metrics. This study introduces NACHOS (Nested and Automated Cross-validation
and Hyperparameter Optimization using Supercomputing) to reduce and quantify
the variance of test performance metrics of deep learning models. NACHOS
integrates Nested Cross-Validation (NCV) and Automated Hyperparameter
Optimization (AHPO) within a parallelized high-performance computing (HPC)
framework. NACHOS was demonstrated on a chest X-ray repository and an Optical
Coherence Tomography (OCT) dataset under multiple data partitioning schemes.
Beyond performance estimation, DACHOS (Deployment with Automated
Cross-validation and Hyperparameter Optimization using Supercomputing) is
introduced to leverage AHPO and cross-validation to build the final model on
the full dataset, improving expected deployment performance. The findings
underscore the importance of NCV in quantifying and reducing estimation
variance, AHPO in optimizing hyperparameters consistently across test folds,
and HPC in ensuring computational feasibility. By integrating these
methodologies, NACHOS and DACHOS provide a scalable, reproducible, and
trustworthy framework for DL model evaluation and deployment in medical
imaging.
| no_new_dataset | 0.949106 |
2503.08596 | Feiran Wang | Feiran Wang, Jiachen Tao, Junyi Wu, Haoxuan Wang, Bin Duan, Kai Wang,
Zongxin Yang, Yan Yan | X-Field: A Physically Grounded Representation for 3D X-ray
Reconstruction | Project Page: \url{https://brack-wang.github.io/XField/}, Github
Code: \url{https://github.com/Brack-Wang/X-Field} | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | X-ray imaging is indispensable in medical diagnostics, yet its use is tightly
regulated due to potential health risks. To mitigate radiation exposure, recent
research focuses on generating novel views from sparse inputs and
reconstructing Computed Tomography (CT) volumes, borrowing representations from
the 3D reconstruction area. However, these representations originally target
visible light imaging that emphasizes reflection and scattering effects, while
neglecting penetration and attenuation properties of X-ray imaging. In this
paper, we introduce X-Field, the first 3D representation specifically designed
for X-ray imaging, rooted in the energy absorption rates across different
materials. To accurately model diverse materials within internal structures, we
employ 3D ellipsoids with distinct attenuation coefficients. To estimate each
material's energy absorption of X-rays, we devise an efficient path
partitioning algorithm accounting for complex ellipsoid intersections. We
further propose hybrid progressive initialization to refine the geometric
accuracy of X-Filed and incorporate material-based optimization to enhance
model fitting along material boundaries. Experiments show that X-Field achieves
superior visual fidelity on both real-world human organ and synthetic object
datasets, outperforming state-of-the-art methods in X-ray Novel View Synthesis
and CT Reconstruction.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 16:31:56 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Wang",
"Feiran",
""
],
[
"Tao",
"Jiachen",
""
],
[
"Wu",
"Junyi",
""
],
[
"Wang",
"Haoxuan",
""
],
[
"Duan",
"Bin",
""
],
[
"Wang",
"Kai",
""
],
[
"Yang",
"Zongxin",
""
],
[
"Yan",
"Yan",
""
]
]
| TITLE: X-Field: A Physically Grounded Representation for 3D X-ray
Reconstruction
ABSTRACT: X-ray imaging is indispensable in medical diagnostics, yet its use is tightly
regulated due to potential health risks. To mitigate radiation exposure, recent
research focuses on generating novel views from sparse inputs and
reconstructing Computed Tomography (CT) volumes, borrowing representations from
the 3D reconstruction area. However, these representations originally target
visible light imaging that emphasizes reflection and scattering effects, while
neglecting penetration and attenuation properties of X-ray imaging. In this
paper, we introduce X-Field, the first 3D representation specifically designed
for X-ray imaging, rooted in the energy absorption rates across different
materials. To accurately model diverse materials within internal structures, we
employ 3D ellipsoids with distinct attenuation coefficients. To estimate each
material's energy absorption of X-rays, we devise an efficient path
partitioning algorithm accounting for complex ellipsoid intersections. We
further propose hybrid progressive initialization to refine the geometric
accuracy of X-Filed and incorporate material-based optimization to enhance
model fitting along material boundaries. Experiments show that X-Field achieves
superior visual fidelity on both real-world human organ and synthetic object
datasets, outperforming state-of-the-art methods in X-ray Novel View Synthesis
and CT Reconstruction.
| no_new_dataset | 0.949201 |
2503.08601 | Du\v{s}an Mali\'c | Du\v{s}an Mali\'c, Christian Fruhwirth-Reisinger, Samuel Schulter,
Horst Possegger | LiSu: A Dataset and Method for LiDAR Surface Normal Estimation | Accepted at CVPR 2025 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | While surface normals are widely used to analyse 3D scene geometry, surface
normal estimation from LiDAR point clouds remains severely underexplored. This
is caused by the lack of large-scale annotated datasets on the one hand, and
lack of methods that can robustly handle the sparse and often noisy LiDAR data
in a reasonable time on the other hand. We address these limitations using a
traffic simulation engine and present LiSu, the first large-scale, synthetic
LiDAR point cloud dataset with ground truth surface normal annotations,
eliminating the need for tedious manual labeling. Additionally, we propose a
novel method that exploits the spatiotemporal characteristics of autonomous
driving data to enhance surface normal estimation accuracy. By incorporating
two regularization terms, we enforce spatial consistency among neighboring
points and temporal smoothness across consecutive LiDAR frames. These
regularizers are particularly effective in self-training settings, where they
mitigate the impact of noisy pseudo-labels, enabling robust real-world
deployment. We demonstrate the effectiveness of our method on LiSu, achieving
state-of-the-art performance in LiDAR surface normal estimation. Moreover, we
showcase its full potential in addressing the challenging task of
synthetic-to-real domain adaptation, leading to improved neural surface
reconstruction on real-world data.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 16:35:22 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Malić",
"Dušan",
""
],
[
"Fruhwirth-Reisinger",
"Christian",
""
],
[
"Schulter",
"Samuel",
""
],
[
"Possegger",
"Horst",
""
]
]
| TITLE: LiSu: A Dataset and Method for LiDAR Surface Normal Estimation
ABSTRACT: While surface normals are widely used to analyse 3D scene geometry, surface
normal estimation from LiDAR point clouds remains severely underexplored. This
is caused by the lack of large-scale annotated datasets on the one hand, and
lack of methods that can robustly handle the sparse and often noisy LiDAR data
in a reasonable time on the other hand. We address these limitations using a
traffic simulation engine and present LiSu, the first large-scale, synthetic
LiDAR point cloud dataset with ground truth surface normal annotations,
eliminating the need for tedious manual labeling. Additionally, we propose a
novel method that exploits the spatiotemporal characteristics of autonomous
driving data to enhance surface normal estimation accuracy. By incorporating
two regularization terms, we enforce spatial consistency among neighboring
points and temporal smoothness across consecutive LiDAR frames. These
regularizers are particularly effective in self-training settings, where they
mitigate the impact of noisy pseudo-labels, enabling robust real-world
deployment. We demonstrate the effectiveness of our method on LiSu, achieving
state-of-the-art performance in LiDAR surface normal estimation. Moreover, we
showcase its full potential in addressing the challenging task of
synthetic-to-real domain adaptation, leading to improved neural surface
reconstruction on real-world data.
| no_new_dataset | 0.906031 |
2503.08603 | R\"uveyda Yilmaz | R\"uveyda Yilmaz, Zhu Chen, Yuli Wu and Johannes Stegmaier | CellStyle: Improved Zero-Shot Cell Segmentation via Style Transfer | null | null | null | null | cs.LG cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Cell microscopy data are abundant; however, corresponding segmentation
annotations remain scarce. Moreover, variations in cell types, imaging devices,
and staining techniques introduce significant domain gaps between datasets. As
a result, even large, pretrained segmentation models trained on diverse
datasets (source datasets) struggle to generalize to unseen datasets (target
datasets). To overcome this generalization problem, we propose CellStyle, which
improves the segmentation quality of such models without requiring labels for
the target dataset, thereby enabling zero-shot adaptation. CellStyle transfers
the attributes of an unannotated target dataset, such as texture, color, and
noise, to the annotated source dataset. This transfer is performed while
preserving the cell shapes of the source images, ensuring that the existing
source annotations can still be used while maintaining the visual
characteristics of the target dataset. The styled synthetic images with the
existing annotations enable the finetuning of a generalist segmentation model
for application to the unannotated target data. We demonstrate that CellStyle
significantly improves zero-shot cell segmentation performance across diverse
datasets by finetuning multiple segmentation models on the style-transferred
data. The code will be made publicly available.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 16:39:09 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Yilmaz",
"Rüveyda",
""
],
[
"Chen",
"Zhu",
""
],
[
"Wu",
"Yuli",
""
],
[
"Stegmaier",
"Johannes",
""
]
]
| TITLE: CellStyle: Improved Zero-Shot Cell Segmentation via Style Transfer
ABSTRACT: Cell microscopy data are abundant; however, corresponding segmentation
annotations remain scarce. Moreover, variations in cell types, imaging devices,
and staining techniques introduce significant domain gaps between datasets. As
a result, even large, pretrained segmentation models trained on diverse
datasets (source datasets) struggle to generalize to unseen datasets (target
datasets). To overcome this generalization problem, we propose CellStyle, which
improves the segmentation quality of such models without requiring labels for
the target dataset, thereby enabling zero-shot adaptation. CellStyle transfers
the attributes of an unannotated target dataset, such as texture, color, and
noise, to the annotated source dataset. This transfer is performed while
preserving the cell shapes of the source images, ensuring that the existing
source annotations can still be used while maintaining the visual
characteristics of the target dataset. The styled synthetic images with the
existing annotations enable the finetuning of a generalist segmentation model
for application to the unannotated target data. We demonstrate that CellStyle
significantly improves zero-shot cell segmentation performance across diverse
datasets by finetuning multiple segmentation models on the style-transferred
data. The code will be made publicly available.
| no_new_dataset | 0.9549 |
2503.08604 | Dongping Li | Dongping Li, Tielong Cai, Tianci Tang, Wenhao Chai, Katherine Rose
Driggs-Campbell, Gaoang Wang | EMMOE: A Comprehensive Benchmark for Embodied Mobile Manipulation in
Open Environments | null | null | null | null | cs.RO cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Developing autonomous home robots controlled by natural language has long
been a pursuit of human. While advancements in large language models (LLMs) and
embodied intelligence make this goal closer, several challenges persist: the
lack of a unified benchmark for more complex robot tasks, limited evaluation
methods and metrics, data incompatibility between LLMs and mobile manipulation
trajectories. To address these issues, we introduce Embodied Mobile
Manipulation in Open Environments (EMMOE), which requires agents to interpret
user instructions and execute long-horizon everyday tasks in continuous space.
EMMOE seamlessly integrates high-level and low-level embodied tasks into a
unified framework, along with three new metrics for more diverse assessment.
Additionally, we collect EMMOE-100, which features in various task attributes,
detailed process annotations, re-plans after failures, and two sub-datasets for
LLM training. Furthermore, we design HomieBot, a sophisticated agent system
consists of LLM with Direct Preference Optimization (DPO), light weighted
navigation and manipulation models, and multiple error detection mechanisms.
Finally, we demonstrate HomieBot's performance and the evaluation of different
models and policies.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 16:42:36 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Li",
"Dongping",
""
],
[
"Cai",
"Tielong",
""
],
[
"Tang",
"Tianci",
""
],
[
"Chai",
"Wenhao",
""
],
[
"Driggs-Campbell",
"Katherine Rose",
""
],
[
"Wang",
"Gaoang",
""
]
]
| TITLE: EMMOE: A Comprehensive Benchmark for Embodied Mobile Manipulation in
Open Environments
ABSTRACT: Developing autonomous home robots controlled by natural language has long
been a pursuit of human. While advancements in large language models (LLMs) and
embodied intelligence make this goal closer, several challenges persist: the
lack of a unified benchmark for more complex robot tasks, limited evaluation
methods and metrics, data incompatibility between LLMs and mobile manipulation
trajectories. To address these issues, we introduce Embodied Mobile
Manipulation in Open Environments (EMMOE), which requires agents to interpret
user instructions and execute long-horizon everyday tasks in continuous space.
EMMOE seamlessly integrates high-level and low-level embodied tasks into a
unified framework, along with three new metrics for more diverse assessment.
Additionally, we collect EMMOE-100, which features in various task attributes,
detailed process annotations, re-plans after failures, and two sub-datasets for
LLM training. Furthermore, we design HomieBot, a sophisticated agent system
consists of LLM with Direct Preference Optimization (DPO), light weighted
navigation and manipulation models, and multiple error detection mechanisms.
Finally, we demonstrate HomieBot's performance and the evaluation of different
models and policies.
| new_dataset | 0.95275 |
2503.08612 | Erkang Cheng | Yingqi Tang, Zhuoran Xu, Zhaotie Meng, Erkang Cheng | HiP-AD: Hierarchical and Multi-Granularity Planning with Deformable
Attention for Autonomous Driving in a Single Decoder | null | null | null | null | cs.RO cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Although end-to-end autonomous driving (E2E-AD) technologies have made
significant progress in recent years, there remains an unsatisfactory
performance on closed-loop evaluation. The potential of leveraging planning in
query design and interaction has not yet been fully explored. In this paper, we
introduce a multi-granularity planning query representation that integrates
heterogeneous waypoints, including spatial, temporal, and driving-style
waypoints across various sampling patterns. It provides additional supervision
for trajectory prediction, enhancing precise closed-loop control for the ego
vehicle. Additionally, we explicitly utilize the geometric properties of
planning trajectories to effectively retrieve relevant image features based on
physical locations using deformable attention. By combining these strategies,
we propose a novel end-to-end autonomous driving framework, termed HiP-AD,
which simultaneously performs perception, prediction, and planning within a
unified decoder. HiP-AD enables comprehensive interaction by allowing planning
queries to iteratively interact with perception queries in the BEV space while
dynamically extracting image features from perspective views. Experiments
demonstrate that HiP-AD outperforms all existing end-to-end autonomous driving
methods on the closed-loop benchmark Bench2Drive and achieves competitive
performance on the real-world dataset nuScenes.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 16:52:45 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Tang",
"Yingqi",
""
],
[
"Xu",
"Zhuoran",
""
],
[
"Meng",
"Zhaotie",
""
],
[
"Cheng",
"Erkang",
""
]
]
| TITLE: HiP-AD: Hierarchical and Multi-Granularity Planning with Deformable
Attention for Autonomous Driving in a Single Decoder
ABSTRACT: Although end-to-end autonomous driving (E2E-AD) technologies have made
significant progress in recent years, there remains an unsatisfactory
performance on closed-loop evaluation. The potential of leveraging planning in
query design and interaction has not yet been fully explored. In this paper, we
introduce a multi-granularity planning query representation that integrates
heterogeneous waypoints, including spatial, temporal, and driving-style
waypoints across various sampling patterns. It provides additional supervision
for trajectory prediction, enhancing precise closed-loop control for the ego
vehicle. Additionally, we explicitly utilize the geometric properties of
planning trajectories to effectively retrieve relevant image features based on
physical locations using deformable attention. By combining these strategies,
we propose a novel end-to-end autonomous driving framework, termed HiP-AD,
which simultaneously performs perception, prediction, and planning within a
unified decoder. HiP-AD enables comprehensive interaction by allowing planning
queries to iteratively interact with perception queries in the BEV space while
dynamically extracting image features from perspective views. Experiments
demonstrate that HiP-AD outperforms all existing end-to-end autonomous driving
methods on the closed-loop benchmark Bench2Drive and achieves competitive
performance on the real-world dataset nuScenes.
| no_new_dataset | 0.943815 |
2503.08619 | Xianfeng Wu | Xianfeng Wu, Yajing Bai, Haoze Zheng, Harold Haodong Chen, Yexin Liu,
Zihao Wang, Xuran Ma, Wen-Jie Shu, Xianzu Wu, Harry Yang, Ser-Nam Lim | LightGen: Efficient Image Generation through Knowledge Distillation and
Direct Preference Optimization | Code: https://github.com/XianfengWu01/LightGen | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advances in text-to-image generation have primarily relied on
extensive datasets and parameter-heavy architectures. These requirements
severely limit accessibility for researchers and practitioners who lack
substantial computational resources. In this paper, we introduce \model, an
efficient training paradigm for image generation models that uses knowledge
distillation (KD) and Direct Preference Optimization (DPO). Drawing inspiration
from the success of data KD techniques widely adopted in Multi-Modal Large
Language Models (MLLMs), LightGen distills knowledge from state-of-the-art
(SOTA) text-to-image models into a compact Masked Autoregressive (MAR)
architecture with only $0.7B$ parameters. Using a compact synthetic dataset of
just $2M$ high-quality images generated from varied captions, we demonstrate
that data diversity significantly outweighs data volume in determining model
performance. This strategy dramatically reduces computational demands and
reduces pre-training time from potentially thousands of GPU-days to merely 88
GPU-days. Furthermore, to address the inherent shortcomings of synthetic data,
particularly poor high-frequency details and spatial inaccuracies, we integrate
the DPO technique that refines image fidelity and positional accuracy.
Comprehensive experiments confirm that LightGen achieves image generation
quality comparable to SOTA models while significantly reducing computational
resources and expanding accessibility for resource-constrained environments.
Code is available at https://github.com/XianfengWu01/LightGen
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 16:58:02 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Wu",
"Xianfeng",
""
],
[
"Bai",
"Yajing",
""
],
[
"Zheng",
"Haoze",
""
],
[
"Chen",
"Harold Haodong",
""
],
[
"Liu",
"Yexin",
""
],
[
"Wang",
"Zihao",
""
],
[
"Ma",
"Xuran",
""
],
[
"Shu",
"Wen-Jie",
""
],
[
"Wu",
"Xianzu",
""
],
[
"Yang",
"Harry",
""
],
[
"Lim",
"Ser-Nam",
""
]
]
| TITLE: LightGen: Efficient Image Generation through Knowledge Distillation and
Direct Preference Optimization
ABSTRACT: Recent advances in text-to-image generation have primarily relied on
extensive datasets and parameter-heavy architectures. These requirements
severely limit accessibility for researchers and practitioners who lack
substantial computational resources. In this paper, we introduce \model, an
efficient training paradigm for image generation models that uses knowledge
distillation (KD) and Direct Preference Optimization (DPO). Drawing inspiration
from the success of data KD techniques widely adopted in Multi-Modal Large
Language Models (MLLMs), LightGen distills knowledge from state-of-the-art
(SOTA) text-to-image models into a compact Masked Autoregressive (MAR)
architecture with only $0.7B$ parameters. Using a compact synthetic dataset of
just $2M$ high-quality images generated from varied captions, we demonstrate
that data diversity significantly outweighs data volume in determining model
performance. This strategy dramatically reduces computational demands and
reduces pre-training time from potentially thousands of GPU-days to merely 88
GPU-days. Furthermore, to address the inherent shortcomings of synthetic data,
particularly poor high-frequency details and spatial inaccuracies, we integrate
the DPO technique that refines image fidelity and positional accuracy.
Comprehensive experiments confirm that LightGen achieves image generation
quality comparable to SOTA models while significantly reducing computational
resources and expanding accessibility for resource-constrained environments.
Code is available at https://github.com/XianfengWu01/LightGen
| no_new_dataset | 0.940517 |
2503.08622 | Apan Dastider | Apan Dastider, Hao Fang and Mingjie Lin | Cross-Embodiment Robotic Manipulation Synthesis via Guided
Demonstrations through CycleVAE and Human Behavior Transformer | Under Review in IROS 2025 | null | null | null | cs.RO | http://creativecommons.org/licenses/by/4.0/ | Cross-embodiment robotic manipulation synthesis for complicated tasks is
challenging, partially due to the scarcity of paired cross-embodiment datasets
and the impediment of designing intricate controllers. Inspired by robotic
learning via guided human expert demonstration, we here propose a novel
cross-embodiment robotic manipulation algorithm via CycleVAE and human behavior
transformer. First, we utilize unsupervised CycleVAE together with a
bidirectional subspace alignment algorithm to align latent motion sequences
between cross-embodiments. Second, we propose a casual human behavior
transformer design to learn the intrinsic motion dynamics of human expert
demonstrations. During the test case, we leverage the proposed transformer for
the human expert demonstration generation, which will be aligned using CycleVAE
for the final human-robotic manipulation synthesis. We validated our proposed
algorithm through extensive experiments using a dexterous robotic manipulator
with the robotic hand. Our results successfully generate smooth trajectories
across intricate tasks, outperforming prior learning-based robotic motion
planning algorithms. These results have implications for performing
unsupervised cross-embodiment alignment and future autonomous robotics design.
Complete video demonstrations of our experiments can be found in
https://sites.google.com/view/humanrobots/home.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 17:02:08 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Dastider",
"Apan",
""
],
[
"Fang",
"Hao",
""
],
[
"Lin",
"Mingjie",
""
]
]
| TITLE: Cross-Embodiment Robotic Manipulation Synthesis via Guided
Demonstrations through CycleVAE and Human Behavior Transformer
ABSTRACT: Cross-embodiment robotic manipulation synthesis for complicated tasks is
challenging, partially due to the scarcity of paired cross-embodiment datasets
and the impediment of designing intricate controllers. Inspired by robotic
learning via guided human expert demonstration, we here propose a novel
cross-embodiment robotic manipulation algorithm via CycleVAE and human behavior
transformer. First, we utilize unsupervised CycleVAE together with a
bidirectional subspace alignment algorithm to align latent motion sequences
between cross-embodiments. Second, we propose a casual human behavior
transformer design to learn the intrinsic motion dynamics of human expert
demonstrations. During the test case, we leverage the proposed transformer for
the human expert demonstration generation, which will be aligned using CycleVAE
for the final human-robotic manipulation synthesis. We validated our proposed
algorithm through extensive experiments using a dexterous robotic manipulator
with the robotic hand. Our results successfully generate smooth trajectories
across intricate tasks, outperforming prior learning-based robotic motion
planning algorithms. These results have implications for performing
unsupervised cross-embodiment alignment and future autonomous robotics design.
Complete video demonstrations of our experiments can be found in
https://sites.google.com/view/humanrobots/home.
| no_new_dataset | 0.950273 |
2503.08639 | Du\v{s}an Mali\'c | Du\v{s}an Mali\'c, Christian Fruhwirth-Reisinger, Samuel Schulter,
Horst Possegger | GBlobs: Explicit Local Structure via Gaussian Blobs for Improved
Cross-Domain LiDAR-based 3D Object Detection | Accepted at CVPR 2025 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | LiDAR-based 3D detectors need large datasets for training, yet they struggle
to generalize to novel domains. Domain Generalization (DG) aims to mitigate
this by training detectors that are invariant to such domain shifts. Current DG
approaches exclusively rely on global geometric features (point cloud Cartesian
coordinates) as input features. Over-reliance on these global geometric
features can, however, cause 3D detectors to prioritize object location and
absolute position, resulting in poor cross-domain performance. To mitigate
this, we propose to exploit explicit local point cloud structure for DG, in
particular by encoding point cloud neighborhoods with Gaussian blobs, GBlobs.
Our proposed formulation is highly efficient and requires no additional
parameters. Without any bells and whistles, simply by integrating GBlobs in
existing detectors, we beat the current state-of-the-art in challenging
single-source DG benchmarks by over 21 mAP (Waymo->KITTI), 13 mAP
(KITTI->Waymo), and 12 mAP (nuScenes->KITTI), without sacrificing in-domain
performance. Additionally, GBlobs demonstrate exceptional performance in
multi-source DG, surpassing the current state-of-the-art by 17, 12, and 5 mAP
on Waymo, KITTI, and ONCE, respectively.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 17:29:56 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Malić",
"Dušan",
""
],
[
"Fruhwirth-Reisinger",
"Christian",
""
],
[
"Schulter",
"Samuel",
""
],
[
"Possegger",
"Horst",
""
]
]
| TITLE: GBlobs: Explicit Local Structure via Gaussian Blobs for Improved
Cross-Domain LiDAR-based 3D Object Detection
ABSTRACT: LiDAR-based 3D detectors need large datasets for training, yet they struggle
to generalize to novel domains. Domain Generalization (DG) aims to mitigate
this by training detectors that are invariant to such domain shifts. Current DG
approaches exclusively rely on global geometric features (point cloud Cartesian
coordinates) as input features. Over-reliance on these global geometric
features can, however, cause 3D detectors to prioritize object location and
absolute position, resulting in poor cross-domain performance. To mitigate
this, we propose to exploit explicit local point cloud structure for DG, in
particular by encoding point cloud neighborhoods with Gaussian blobs, GBlobs.
Our proposed formulation is highly efficient and requires no additional
parameters. Without any bells and whistles, simply by integrating GBlobs in
existing detectors, we beat the current state-of-the-art in challenging
single-source DG benchmarks by over 21 mAP (Waymo->KITTI), 13 mAP
(KITTI->Waymo), and 12 mAP (nuScenes->KITTI), without sacrificing in-domain
performance. Additionally, GBlobs demonstrate exceptional performance in
multi-source DG, surpassing the current state-of-the-art by 17, 12, and 5 mAP
on Waymo, KITTI, and ONCE, respectively.
| no_new_dataset | 0.947769 |
2503.08642 | Zecheng Zhang | Zecheng Zhang, Hao Liu, Wenjing Liao, Guang Lin | Coefficient-to-Basis Network: A Fine-Tunable Operator Learning Framework
for Inverse Problems with Adaptive Discretizations and Theoretical Guarantees | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | We propose a Coefficient-to-Basis Network (C2BNet), a novel framework for
solving inverse problems within the operator learning paradigm. C2BNet
efficiently adapts to different discretizations through fine-tuning, using a
pre-trained model to significantly reduce computational cost while maintaining
high accuracy. Unlike traditional approaches that require retraining from
scratch for new discretizations, our method enables seamless adaptation without
sacrificing predictive performance. Furthermore, we establish theoretical
approximation and generalization error bounds for C2BNet by exploiting
low-dimensional structures in the underlying datasets. Our analysis
demonstrates that C2BNet adapts to low-dimensional structures without relying
on explicit encoding mechanisms, highlighting its robustness and efficiency. To
validate our theoretical findings, we conducted extensive numerical experiments
that showcase the superior performance of C2BNet on several inverse problems.
The results confirm that C2BNet effectively balances computational efficiency
and accuracy, making it a promising tool to solve inverse problems in
scientific computing and engineering applications.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 17:34:38 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Zhang",
"Zecheng",
""
],
[
"Liu",
"Hao",
""
],
[
"Liao",
"Wenjing",
""
],
[
"Lin",
"Guang",
""
]
]
| TITLE: Coefficient-to-Basis Network: A Fine-Tunable Operator Learning Framework
for Inverse Problems with Adaptive Discretizations and Theoretical Guarantees
ABSTRACT: We propose a Coefficient-to-Basis Network (C2BNet), a novel framework for
solving inverse problems within the operator learning paradigm. C2BNet
efficiently adapts to different discretizations through fine-tuning, using a
pre-trained model to significantly reduce computational cost while maintaining
high accuracy. Unlike traditional approaches that require retraining from
scratch for new discretizations, our method enables seamless adaptation without
sacrificing predictive performance. Furthermore, we establish theoretical
approximation and generalization error bounds for C2BNet by exploiting
low-dimensional structures in the underlying datasets. Our analysis
demonstrates that C2BNet adapts to low-dimensional structures without relying
on explicit encoding mechanisms, highlighting its robustness and efficiency. To
validate our theoretical findings, we conducted extensive numerical experiments
that showcase the superior performance of C2BNet on several inverse problems.
The results confirm that C2BNet effectively balances computational efficiency
and accuracy, making it a promising tool to solve inverse problems in
scientific computing and engineering applications.
| no_new_dataset | 0.95018 |
2503.08650 | Zhenchen Wan | Zhenchen Wan, Yanwu xu, Dongting Hu, Weilun Cheng, Tianxi Chen,
Zhaoqing Wang, Feng Liu, Tongliang Liu, Mingming Gong | MF-VITON: High-Fidelity Mask-Free Virtual Try-On with Minimal Input | The project page is available at:
https://zhenchenwan.github.io/MF-VITON/ | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Recent advancements in Virtual Try-On (VITON) have significantly improved
image realism and garment detail preservation, driven by powerful text-to-image
(T2I) diffusion models. However, existing methods often rely on user-provided
masks, introducing complexity and performance degradation due to imperfect
inputs, as shown in Fig.1(a). To address this, we propose a Mask-Free VITON
(MF-VITON) framework that achieves realistic VITON using only a single person
image and a target garment, eliminating the requirement for auxiliary masks.
Our approach introduces a novel two-stage pipeline: (1) We leverage existing
Mask-based VITON models to synthesize a high-quality dataset. This dataset
contains diverse, realistic pairs of person images and corresponding garments,
augmented with varied backgrounds to mimic real-world scenarios. (2) The
pre-trained Mask-based model is fine-tuned on the generated dataset, enabling
garment transfer without mask dependencies. This stage simplifies the input
requirements while preserving garment texture and shape fidelity. Our framework
achieves state-of-the-art (SOTA) performance regarding garment transfer
accuracy and visual realism. Notably, the proposed Mask-Free model
significantly outperforms existing Mask-based approaches, setting a new
benchmark and demonstrating a substantial lead over previous approaches. For
more details, visit our project page: https://zhenchenwan.github.io/MF-VITON/.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 17:40:59 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Wan",
"Zhenchen",
""
],
[
"xu",
"Yanwu",
""
],
[
"Hu",
"Dongting",
""
],
[
"Cheng",
"Weilun",
""
],
[
"Chen",
"Tianxi",
""
],
[
"Wang",
"Zhaoqing",
""
],
[
"Liu",
"Feng",
""
],
[
"Liu",
"Tongliang",
""
],
[
"Gong",
"Mingming",
""
]
]
| TITLE: MF-VITON: High-Fidelity Mask-Free Virtual Try-On with Minimal Input
ABSTRACT: Recent advancements in Virtual Try-On (VITON) have significantly improved
image realism and garment detail preservation, driven by powerful text-to-image
(T2I) diffusion models. However, existing methods often rely on user-provided
masks, introducing complexity and performance degradation due to imperfect
inputs, as shown in Fig.1(a). To address this, we propose a Mask-Free VITON
(MF-VITON) framework that achieves realistic VITON using only a single person
image and a target garment, eliminating the requirement for auxiliary masks.
Our approach introduces a novel two-stage pipeline: (1) We leverage existing
Mask-based VITON models to synthesize a high-quality dataset. This dataset
contains diverse, realistic pairs of person images and corresponding garments,
augmented with varied backgrounds to mimic real-world scenarios. (2) The
pre-trained Mask-based model is fine-tuned on the generated dataset, enabling
garment transfer without mask dependencies. This stage simplifies the input
requirements while preserving garment texture and shape fidelity. Our framework
achieves state-of-the-art (SOTA) performance regarding garment transfer
accuracy and visual realism. Notably, the proposed Mask-Free model
significantly outperforms existing Mask-based approaches, setting a new
benchmark and demonstrating a substantial lead over previous approaches. For
more details, visit our project page: https://zhenchenwan.github.io/MF-VITON/.
| new_dataset | 0.967472 |
2503.08652 | Wei Chen | Wei Chen, Qiang Qiu | Extra Clients at No Extra Cost: Overcome Data Heterogeneity in Federated
Learning with Filter Decomposition | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Data heterogeneity is one of the major challenges in federated learning (FL),
which results in substantial client variance and slow convergence. In this
study, we propose a novel solution: decomposing a convolutional filter in FL
into a linear combination of filter subspace elements, i.e., filter atoms. This
simple technique transforms global filter aggregation in FL into aggregating
filter atoms and their atom coefficients. The key advantage here involves
mathematically generating numerous cross-terms by expanding the product of two
weighted sums from filter atom and atom coefficient. These cross-terms
effectively emulate many additional latent clients, significantly reducing
model variance, which is validated by our theoretical analysis and empirical
observation. Furthermore, our method permits different training schemes for
filter atoms and atom coefficients for highly adaptive model personalization
and communication efficiency. Empirical results on benchmark datasets
demonstrate that our filter decomposition technique substantially improves the
accuracy of FL methods, confirming its efficacy in addressing data
heterogeneity.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 17:42:36 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Chen",
"Wei",
""
],
[
"Qiu",
"Qiang",
""
]
]
| TITLE: Extra Clients at No Extra Cost: Overcome Data Heterogeneity in Federated
Learning with Filter Decomposition
ABSTRACT: Data heterogeneity is one of the major challenges in federated learning (FL),
which results in substantial client variance and slow convergence. In this
study, we propose a novel solution: decomposing a convolutional filter in FL
into a linear combination of filter subspace elements, i.e., filter atoms. This
simple technique transforms global filter aggregation in FL into aggregating
filter atoms and their atom coefficients. The key advantage here involves
mathematically generating numerous cross-terms by expanding the product of two
weighted sums from filter atom and atom coefficient. These cross-terms
effectively emulate many additional latent clients, significantly reducing
model variance, which is validated by our theoretical analysis and empirical
observation. Furthermore, our method permits different training schemes for
filter atoms and atom coefficients for highly adaptive model personalization
and communication efficiency. Empirical results on benchmark datasets
demonstrate that our filter decomposition technique substantially improves the
accuracy of FL methods, confirming its efficacy in addressing data
heterogeneity.
| no_new_dataset | 0.953101 |
2503.08663 | Pierre Sermanet | Pierre Sermanet, Anirudha Majumdar, Alex Irpan, Dmitry Kalashnikov,
Vikas Sindhwani | Generating Robot Constitutions & Benchmarks for Semantic Safety | null | null | null | null | cs.RO cs.AI cs.CV cs.CY cs.HC | http://creativecommons.org/licenses/by/4.0/ | Until recently, robotics safety research was predominantly about collision
avoidance and hazard reduction in the immediate vicinity of a robot. Since the
advent of large vision and language models (VLMs), robots are now also capable
of higher-level semantic scene understanding and natural language interactions
with humans. Despite their known vulnerabilities (e.g. hallucinations or
jail-breaking), VLMs are being handed control of robots capable of physical
contact with the real world. This can lead to dangerous behaviors, making
semantic safety for robots a matter of immediate concern. Our contributions in
this paper are two fold: first, to address these emerging risks, we release the
ASIMOV Benchmark, a large-scale and comprehensive collection of datasets for
evaluating and improving semantic safety of foundation models serving as robot
brains. Our data generation recipe is highly scalable: by leveraging text and
image generation techniques, we generate undesirable situations from real-world
visual scenes and human injury reports from hospitals. Secondly, we develop a
framework to automatically generate robot constitutions from real-world data to
steer a robot's behavior using Constitutional AI mechanisms. We propose a novel
auto-amending process that is able to introduce nuances in written rules of
behavior; this can lead to increased alignment with human preferences on
behavior desirability and safety. We explore trade-offs between generality and
specificity across a diverse set of constitutions of different lengths, and
demonstrate that a robot is able to effectively reject unconstitutional
actions. We measure a top alignment rate of 84.3% on the ASIMOV Benchmark using
generated constitutions, outperforming no-constitution baselines and
human-written constitutions. Data is available at asimov-benchmark.github.io
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 17:50:47 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Sermanet",
"Pierre",
""
],
[
"Majumdar",
"Anirudha",
""
],
[
"Irpan",
"Alex",
""
],
[
"Kalashnikov",
"Dmitry",
""
],
[
"Sindhwani",
"Vikas",
""
]
]
| TITLE: Generating Robot Constitutions & Benchmarks for Semantic Safety
ABSTRACT: Until recently, robotics safety research was predominantly about collision
avoidance and hazard reduction in the immediate vicinity of a robot. Since the
advent of large vision and language models (VLMs), robots are now also capable
of higher-level semantic scene understanding and natural language interactions
with humans. Despite their known vulnerabilities (e.g. hallucinations or
jail-breaking), VLMs are being handed control of robots capable of physical
contact with the real world. This can lead to dangerous behaviors, making
semantic safety for robots a matter of immediate concern. Our contributions in
this paper are two fold: first, to address these emerging risks, we release the
ASIMOV Benchmark, a large-scale and comprehensive collection of datasets for
evaluating and improving semantic safety of foundation models serving as robot
brains. Our data generation recipe is highly scalable: by leveraging text and
image generation techniques, we generate undesirable situations from real-world
visual scenes and human injury reports from hospitals. Secondly, we develop a
framework to automatically generate robot constitutions from real-world data to
steer a robot's behavior using Constitutional AI mechanisms. We propose a novel
auto-amending process that is able to introduce nuances in written rules of
behavior; this can lead to increased alignment with human preferences on
behavior desirability and safety. We explore trade-offs between generality and
specificity across a diverse set of constitutions of different lengths, and
demonstrate that a robot is able to effectively reject unconstitutional
actions. We measure a top alignment rate of 84.3% on the ASIMOV Benchmark using
generated constitutions, outperforming no-constitution baselines and
human-written constitutions. Data is available at asimov-benchmark.github.io
| new_dataset | 0.767429 |
2503.08674 | Tobias Kreiman | Tobias Kreiman and Aditi S. Krishnapriyan | Understanding and Mitigating Distribution Shifts For Machine Learning
Force Fields | null | null | null | null | cs.LG cond-mat.mtrl-sci physics.chem-ph q-bio.BM | http://creativecommons.org/licenses/by/4.0/ | Machine Learning Force Fields (MLFFs) are a promising alternative to
expensive ab initio quantum mechanical molecular simulations. Given the
diversity of chemical spaces that are of interest and the cost of generating
new data, it is important to understand how MLFFs generalize beyond their
training distributions. In order to characterize and better understand
distribution shifts in MLFFs, we conduct diagnostic experiments on chemical
datasets, revealing common shifts that pose significant challenges, even for
large foundation models trained on extensive data. Based on these observations,
we hypothesize that current supervised training methods inadequately regularize
MLFFs, resulting in overfitting and learning poor representations of
out-of-distribution systems. We then propose two new methods as initial steps
for mitigating distribution shifts for MLFFs. Our methods focus on test-time
refinement strategies that incur minimal computational cost and do not use
expensive ab initio reference labels. The first strategy, based on spectral
graph theory, modifies the edges of test graphs to align with graph structures
seen during training. Our second strategy improves representations for
out-of-distribution systems at test-time by taking gradient steps using an
auxiliary objective, such as a cheap physical prior. Our test-time refinement
strategies significantly reduce errors on out-of-distribution systems,
suggesting that MLFFs are capable of and can move towards modeling diverse
chemical spaces, but are not being effectively trained to do so. Our
experiments establish clear benchmarks for evaluating the generalization
capabilities of the next generation of MLFFs. Our code is available at
https://tkreiman.github.io/projects/mlff_distribution_shifts/.
| [
{
"version": "v1",
"created": "Tue, 11 Mar 2025 17:54:29 GMT"
}
]
| 2025-03-12T00:00:00 | [
[
"Kreiman",
"Tobias",
""
],
[
"Krishnapriyan",
"Aditi S.",
""
]
]
| TITLE: Understanding and Mitigating Distribution Shifts For Machine Learning
Force Fields
ABSTRACT: Machine Learning Force Fields (MLFFs) are a promising alternative to
expensive ab initio quantum mechanical molecular simulations. Given the
diversity of chemical spaces that are of interest and the cost of generating
new data, it is important to understand how MLFFs generalize beyond their
training distributions. In order to characterize and better understand
distribution shifts in MLFFs, we conduct diagnostic experiments on chemical
datasets, revealing common shifts that pose significant challenges, even for
large foundation models trained on extensive data. Based on these observations,
we hypothesize that current supervised training methods inadequately regularize
MLFFs, resulting in overfitting and learning poor representations of
out-of-distribution systems. We then propose two new methods as initial steps
for mitigating distribution shifts for MLFFs. Our methods focus on test-time
refinement strategies that incur minimal computational cost and do not use
expensive ab initio reference labels. The first strategy, based on spectral
graph theory, modifies the edges of test graphs to align with graph structures
seen during training. Our second strategy improves representations for
out-of-distribution systems at test-time by taking gradient steps using an
auxiliary objective, such as a cheap physical prior. Our test-time refinement
strategies significantly reduce errors on out-of-distribution systems,
suggesting that MLFFs are capable of and can move towards modeling diverse
chemical spaces, but are not being effectively trained to do so. Our
experiments establish clear benchmarks for evaluating the generalization
capabilities of the next generation of MLFFs. Our code is available at
https://tkreiman.github.io/projects/mlff_distribution_shifts/.
| no_new_dataset | 0.951323 |
2105.05717 | Lunchen Xie | Lunchen Xie, Jiaqi Liu, Songtao Lu, Tsung-hui Chang, Qingjiang Shi | An Efficient Learning Framework For Federated XGBoost Using Secret
Sharing And Distributed Optimization | 24 pages, Special issue of ACM Transactions on Intelligent Systems
and Technology | null | 10.1145/3523061 | null | cs.LG cs.AI cs.CR | http://creativecommons.org/licenses/by/4.0/ | XGBoost is one of the most widely used machine learning models in the
industry due to its superior learning accuracy and efficiency. Targeting at
data isolation issues in the big data problems, it is crucial to deploy a
secure and efficient federated XGBoost (FedXGB) model. Existing FedXGB models
either have data leakage issues or are only applicable to the two-party setting
with heavy communication and computation overheads. In this paper, a lossless
multi-party federated XGB learning framework is proposed with a security
guarantee, which reshapes the XGBoost's split criterion calculation process
under a secret sharing setting and solves the leaf weight calculation problem
by leveraging distributed optimization. Remarkably, a thorough analysis of
model security is provided as well, and multiple numerical results showcase the
superiority of the proposed FedXGB compared with the state-of-the-art models on
benchmark datasets.
| [
{
"version": "v1",
"created": "Wed, 12 May 2021 15:04:18 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Xie",
"Lunchen",
""
],
[
"Liu",
"Jiaqi",
""
],
[
"Lu",
"Songtao",
""
],
[
"Chang",
"Tsung-hui",
""
],
[
"Shi",
"Qingjiang",
""
]
]
| TITLE: An Efficient Learning Framework For Federated XGBoost Using Secret
Sharing And Distributed Optimization
ABSTRACT: XGBoost is one of the most widely used machine learning models in the
industry due to its superior learning accuracy and efficiency. Targeting at
data isolation issues in the big data problems, it is crucial to deploy a
secure and efficient federated XGBoost (FedXGB) model. Existing FedXGB models
either have data leakage issues or are only applicable to the two-party setting
with heavy communication and computation overheads. In this paper, a lossless
multi-party federated XGB learning framework is proposed with a security
guarantee, which reshapes the XGBoost's split criterion calculation process
under a secret sharing setting and solves the leaf weight calculation problem
by leveraging distributed optimization. Remarkably, a thorough analysis of
model security is provided as well, and multiple numerical results showcase the
superiority of the proposed FedXGB compared with the state-of-the-art models on
benchmark datasets.
| no_new_dataset | 0.939637 |
2108.08618 | Martijn Pieter Anton Starmans | Martijn P. A. Starmans, Sebastian R. van der Voort, Thomas Phil, Milea
J. M. Timbergen, Melissa Vos, Guillaume A. Padmos, Wouter Kessels, David
Hanff, Dirk J. Grunhagen, Cornelis Verhoef, Stefan Sleijfer, Martin J. van
den Bent, Marion Smits, Roy S. Dwarkasing, Christopher J. Els, Federico
Fiduzi, Geert J. L. H. van Leenders, Anela Blazevic, Johannes Hofland, Tessa
Brabander, Renza A. H. van Gils, Gaston J. H. Franssen, Richard A. Feelders,
Wouter W. de Herder, Florian E. Buisman, Francois E. J. A. Willemssen, Bas
Groot Koerkamp, Lindsay Angus, Astrid A. M. van der Veldt, Ana Rajicic,
Arlette E. Odink, Mitchell Deen, Jose M. Castillo T., Jifke Veenland, Ivo
Schoots, Michel Renckens, Michail Doukas, Rob A. de Man, Jan N. M. IJzermans,
Razvan L. Miclea, Peter B. Vermeulen, Esther E. Bron, Maarten G. Thomeer,
Jacob J. Visser, Wiro J. Niessen, Stefan Klein (for the Alzheimers Disease
Neuroimaging Initiative) | An automated machine learning framework to optimize radiomics model
construction validated on twelve clinical applications | 22 pages, 3 figures, 2 tables, 1 algorithm, 3 supplementary figures,
4 supplementary tables, 1 supplementary algorithm | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Predicting clinical outcomes from medical images using quantitative features
(``radiomics'') requires many method design choices, Currently, in new clinical
applications, finding the optimal radiomics method out of the wide range of
methods relies on a manual, heuristic trial-and-error process. We introduce a
novel automated framework that optimizes radiomics workflow construction per
application by standardizing the radiomics workflow in modular components,
including a large collection of algorithms for each component, and formulating
a combined algorithm selection and hyperparameter optimization problem. To
solve it, we employ automated machine learning through two strategies (random
search and Bayesian optimization) and three ensembling approaches. Results show
that a medium-sized random search and straight-forward ensembling perform
similar to more advanced methods while being more efficient. Validated across
twelve clinical applications, our approach outperforms both a radiomics
baseline and human experts. Concluding, our framework improves and streamlines
radiomics research by fully automatically optimizing radiomics workflow
construction. To facilitate reproducibility, we publicly release six datasets,
software of the method, and code to reproduce this study.
| [
{
"version": "v1",
"created": "Thu, 19 Aug 2021 11:03:54 GMT"
},
{
"version": "v2",
"created": "Fri, 29 Jul 2022 13:36:52 GMT"
},
{
"version": "v3",
"created": "Mon, 10 Mar 2025 12:20:03 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Starmans",
"Martijn P. A.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"van der Voort",
"Sebastian R.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Phil",
"Thomas",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Timbergen",
"Milea J. M.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Vos",
"Melissa",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Padmos",
"Guillaume A.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Kessels",
"Wouter",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Hanff",
"David",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Grunhagen",
"Dirk J.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Verhoef",
"Cornelis",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Sleijfer",
"Stefan",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Bent",
"Martin J. van den",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Smits",
"Marion",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Dwarkasing",
"Roy S.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Els",
"Christopher J.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Fiduzi",
"Federico",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"van Leenders",
"Geert J. L. H.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Blazevic",
"Anela",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Hofland",
"Johannes",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Brabander",
"Tessa",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"van Gils",
"Renza A. H.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Franssen",
"Gaston J. H.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Feelders",
"Richard A.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"de Herder",
"Wouter W.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Buisman",
"Florian E.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Willemssen",
"Francois E. J. A.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Koerkamp",
"Bas Groot",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Angus",
"Lindsay",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"van der Veldt",
"Astrid A. M.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Rajicic",
"Ana",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Odink",
"Arlette E.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Deen",
"Mitchell",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"T.",
"Jose M. Castillo",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Veenland",
"Jifke",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Schoots",
"Ivo",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Renckens",
"Michel",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Doukas",
"Michail",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"de Man",
"Rob A.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"IJzermans",
"Jan N. M.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Miclea",
"Razvan L.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Vermeulen",
"Peter B.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Bron",
"Esther E.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Thomeer",
"Maarten G.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Visser",
"Jacob J.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Niessen",
"Wiro J.",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
],
[
"Klein",
"Stefan",
"",
"for the Alzheimers Disease\n Neuroimaging Initiative"
]
]
| TITLE: An automated machine learning framework to optimize radiomics model
construction validated on twelve clinical applications
ABSTRACT: Predicting clinical outcomes from medical images using quantitative features
(``radiomics'') requires many method design choices, Currently, in new clinical
applications, finding the optimal radiomics method out of the wide range of
methods relies on a manual, heuristic trial-and-error process. We introduce a
novel automated framework that optimizes radiomics workflow construction per
application by standardizing the radiomics workflow in modular components,
including a large collection of algorithms for each component, and formulating
a combined algorithm selection and hyperparameter optimization problem. To
solve it, we employ automated machine learning through two strategies (random
search and Bayesian optimization) and three ensembling approaches. Results show
that a medium-sized random search and straight-forward ensembling perform
similar to more advanced methods while being more efficient. Validated across
twelve clinical applications, our approach outperforms both a radiomics
baseline and human experts. Concluding, our framework improves and streamlines
radiomics research by fully automatically optimizing radiomics workflow
construction. To facilitate reproducibility, we publicly release six datasets,
software of the method, and code to reproduce this study.
| new_dataset | 0.585823 |
2112.11594 | Haoran You | Haoran You, Tong Geng, Yongan Zhang, Ang Li, Yingyan Celine Lin | GCoD: Graph Convolutional Network Acceleration via Dedicated Algorithm
and Accelerator Co-Design | Published as a conference paper at HPCA 2022 | null | null | null | cs.AR cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Graph Convolutional Networks (GCNs) have emerged as the state-of-the-art
graph learning model. However, it can be notoriously challenging to inference
GCNs over large graph datasets, limiting their application to large real-world
graphs and hindering the exploration of deeper and more sophisticated GCN
graphs. This is because real-world graphs can be extremely large and sparse.
Furthermore, the node degree of GCNs tends to follow the power-law distribution
and therefore have highly irregular adjacency matrices, resulting in
prohibitive inefficiencies in both data processing and movement and thus
substantially limiting the achievable GCN acceleration efficiency. To this end,
this paper proposes a GCN algorithm and accelerator Co-Design framework dubbed
GCoD which can largely alleviate the aforementioned GCN irregularity and boost
GCNs' inference efficiency. Specifically, on the algorithm level, GCoD
integrates a split and conquer GCN training strategy that polarizes the graphs
to be either denser or sparser in local neighborhoods without compromising the
model accuracy, resulting in graph adjacency matrices that (mostly) have merely
two levels of workload and enjoys largely enhanced regularity and thus ease of
acceleration. On the hardware level, we further develop a dedicated two-pronged
accelerator with a separated engine to process each of the aforementioned
denser and sparser workloads, further boosting the overall utilization and
acceleration efficiency. Extensive experiments and ablation studies validate
that our GCoD consistently reduces the number of off-chip accesses, leading to
speedups of 15286x, 294x, 7.8x, and 2.5x as compared to CPUs, GPUs, and
prior-art GCN accelerators including HyGCN and AWB-GCN, respectively, while
maintaining or even improving the task accuracy. Codes are available at
https://github.com/RICE-EIC/GCoD.
| [
{
"version": "v1",
"created": "Wed, 22 Dec 2021 00:30:50 GMT"
},
{
"version": "v2",
"created": "Wed, 30 Mar 2022 23:11:07 GMT"
},
{
"version": "v3",
"created": "Sun, 9 Mar 2025 02:58:24 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"You",
"Haoran",
""
],
[
"Geng",
"Tong",
""
],
[
"Zhang",
"Yongan",
""
],
[
"Li",
"Ang",
""
],
[
"Lin",
"Yingyan Celine",
""
]
]
| TITLE: GCoD: Graph Convolutional Network Acceleration via Dedicated Algorithm
and Accelerator Co-Design
ABSTRACT: Graph Convolutional Networks (GCNs) have emerged as the state-of-the-art
graph learning model. However, it can be notoriously challenging to inference
GCNs over large graph datasets, limiting their application to large real-world
graphs and hindering the exploration of deeper and more sophisticated GCN
graphs. This is because real-world graphs can be extremely large and sparse.
Furthermore, the node degree of GCNs tends to follow the power-law distribution
and therefore have highly irregular adjacency matrices, resulting in
prohibitive inefficiencies in both data processing and movement and thus
substantially limiting the achievable GCN acceleration efficiency. To this end,
this paper proposes a GCN algorithm and accelerator Co-Design framework dubbed
GCoD which can largely alleviate the aforementioned GCN irregularity and boost
GCNs' inference efficiency. Specifically, on the algorithm level, GCoD
integrates a split and conquer GCN training strategy that polarizes the graphs
to be either denser or sparser in local neighborhoods without compromising the
model accuracy, resulting in graph adjacency matrices that (mostly) have merely
two levels of workload and enjoys largely enhanced regularity and thus ease of
acceleration. On the hardware level, we further develop a dedicated two-pronged
accelerator with a separated engine to process each of the aforementioned
denser and sparser workloads, further boosting the overall utilization and
acceleration efficiency. Extensive experiments and ablation studies validate
that our GCoD consistently reduces the number of off-chip accesses, leading to
speedups of 15286x, 294x, 7.8x, and 2.5x as compared to CPUs, GPUs, and
prior-art GCN accelerators including HyGCN and AWB-GCN, respectively, while
maintaining or even improving the task accuracy. Codes are available at
https://github.com/RICE-EIC/GCoD.
| no_new_dataset | 0.94474 |
2302.11341 | A. R. Sricharan | Monika Henzinger and A. R. Sricharan and Teresa Anna Steiner | Differentially Private Continual Release of Histograms and Related
Queries | Accepted at AISTATS 2025 | null | null | null | cs.DS cs.CR | http://creativecommons.org/licenses/by-sa/4.0/ | We study privately releasing column sums of a $d$-dimensional table with
entries from a universe $\chi$ undergoing $T$ row updates, called histogram
under continual release. Our mechanisms give better additive
$\ell_\infty$-error than existing mechanisms for a large class of queries and
input streams. Our first contribution is an output-sensitive mechanism in the
insertions-only model ($\chi = \{0,1\}$) for maintaining (i) the histogram or
(ii) queries that do not require maintaining the entire histogram, such as the
maximum or minimum column sum, the median, or any quantiles. The mechanism has
an additive error of $O(d\log^2 (dq^*)+\log T)$ whp, where $q^*$ is the maximum
output value over all time steps on this dataset. The mechanism does not
require $q^*$ as input. This breaks the $\Omega(d \log T)$ bound of prior work
when $q^* \ll T$. Our second contribution is a mechanism for the turnstile
model that admits negative entry updates ($\chi = \{-1, 0,1\}$). This mechanism
has an additive error of $O(d \log^2 (dK) + \log T)$ whp, where $K$ is the
number of times two consecutive data rows differ, and the mechanism does not
require $K$ as input. This is useful when monitoring inputs that only vary
under unusual circumstances. For $d=1$ this gives the first private mechanism
with error $O(\log^2 K + \log T)$ for continual counting in the turnstile
model, improving on the $O(\log^2 n + \log T)$ error bound by Dwork et al.
[ASIACRYPT 2015], where $n$ is the number of ones in the stream, as well as
allowing negative entries, while Dwork et al. [ASIACRYPT 2015] can only handle
nonnegative entries ($\chi=\{0,1\}$).
| [
{
"version": "v1",
"created": "Wed, 22 Feb 2023 12:38:02 GMT"
},
{
"version": "v2",
"created": "Mon, 10 Mar 2025 12:40:22 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Henzinger",
"Monika",
""
],
[
"Sricharan",
"A. R.",
""
],
[
"Steiner",
"Teresa Anna",
""
]
]
| TITLE: Differentially Private Continual Release of Histograms and Related
Queries
ABSTRACT: We study privately releasing column sums of a $d$-dimensional table with
entries from a universe $\chi$ undergoing $T$ row updates, called histogram
under continual release. Our mechanisms give better additive
$\ell_\infty$-error than existing mechanisms for a large class of queries and
input streams. Our first contribution is an output-sensitive mechanism in the
insertions-only model ($\chi = \{0,1\}$) for maintaining (i) the histogram or
(ii) queries that do not require maintaining the entire histogram, such as the
maximum or minimum column sum, the median, or any quantiles. The mechanism has
an additive error of $O(d\log^2 (dq^*)+\log T)$ whp, where $q^*$ is the maximum
output value over all time steps on this dataset. The mechanism does not
require $q^*$ as input. This breaks the $\Omega(d \log T)$ bound of prior work
when $q^* \ll T$. Our second contribution is a mechanism for the turnstile
model that admits negative entry updates ($\chi = \{-1, 0,1\}$). This mechanism
has an additive error of $O(d \log^2 (dK) + \log T)$ whp, where $K$ is the
number of times two consecutive data rows differ, and the mechanism does not
require $K$ as input. This is useful when monitoring inputs that only vary
under unusual circumstances. For $d=1$ this gives the first private mechanism
with error $O(\log^2 K + \log T)$ for continual counting in the turnstile
model, improving on the $O(\log^2 n + \log T)$ error bound by Dwork et al.
[ASIACRYPT 2015], where $n$ is the number of ones in the stream, as well as
allowing negative entries, while Dwork et al. [ASIACRYPT 2015] can only handle
nonnegative entries ($\chi=\{0,1\}$).
| no_new_dataset | 0.939137 |
2306.17184 | Vinoth Nandakumar | Vinoth Nandakumar, Qiang Qu, Peng Mi and Tongliang Liu | State space models can express n-gram languages | Published in "Transactions on Machine Learning Research", 2025 | null | null | null | cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Recent advancements in recurrent neural networks (RNNs) have reinvigorated
interest in their application to natural language processing tasks,
particularly with the development of more efficient and parallelizable variants
known as state space models (SSMs), which have shown competitive performance
against transformer models while maintaining a lower memory footprint. While
RNNs and SSMs (e.g., Mamba) have been empirically more successful than
rule-based systems based on n-gram models, a rigorous theoretical explanation
for this success has not yet been developed, as it is unclear how these models
encode the combinatorial rules that govern the next-word prediction task. In
this paper, we construct state space language models that can solve the
next-word prediction task for languages generated from n-gram rules, thereby
showing that the former are more expressive. Our proof shows how SSMs can
encode n-gram rules using new theoretical results on their memorization
capacity, and demonstrates how their context window can be controlled by
restricting the spectrum of the state transition matrix. We conduct experiments
with a small dataset generated from n-gram rules to show how our framework can
be applied to SSMs and RNNs obtained through gradient-based optimization.
| [
{
"version": "v1",
"created": "Tue, 20 Jun 2023 10:41:23 GMT"
},
{
"version": "v2",
"created": "Sun, 15 Dec 2024 00:24:59 GMT"
},
{
"version": "v3",
"created": "Sun, 9 Mar 2025 06:40:39 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Nandakumar",
"Vinoth",
""
],
[
"Qu",
"Qiang",
""
],
[
"Mi",
"Peng",
""
],
[
"Liu",
"Tongliang",
""
]
]
| TITLE: State space models can express n-gram languages
ABSTRACT: Recent advancements in recurrent neural networks (RNNs) have reinvigorated
interest in their application to natural language processing tasks,
particularly with the development of more efficient and parallelizable variants
known as state space models (SSMs), which have shown competitive performance
against transformer models while maintaining a lower memory footprint. While
RNNs and SSMs (e.g., Mamba) have been empirically more successful than
rule-based systems based on n-gram models, a rigorous theoretical explanation
for this success has not yet been developed, as it is unclear how these models
encode the combinatorial rules that govern the next-word prediction task. In
this paper, we construct state space language models that can solve the
next-word prediction task for languages generated from n-gram rules, thereby
showing that the former are more expressive. Our proof shows how SSMs can
encode n-gram rules using new theoretical results on their memorization
capacity, and demonstrates how their context window can be controlled by
restricting the spectrum of the state transition matrix. We conduct experiments
with a small dataset generated from n-gram rules to show how our framework can
be applied to SSMs and RNNs obtained through gradient-based optimization.
| no_new_dataset | 0.94625 |
2307.03812 | Iksung Kang | Iksung Kang, Qinrong Zhang, Stella X. Yu, Na Ji | Coordinate-based neural representations for computational adaptive
optics in widefield microscopy | 60 pages, 20 figures, 2 tables. Nat Mach Intell (2024) | null | 10.1038/s42256-024-00853-3 | null | eess.IV cs.SY eess.SY physics.optics | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Widefield microscopy is widely used for non-invasive imaging of biological
structures at subcellular resolution. When applied to complex specimen, its
image quality is degraded by sample-induced optical aberration. Adaptive optics
can correct wavefront distortion and restore diffraction-limited resolution but
require wavefront sensing and corrective devices, increasing system complexity
and cost. Here, we describe a self-supervised machine learning algorithm,
CoCoA, that performs joint wavefront estimation and three-dimensional
structural information extraction from a single input 3D image stack without
the need for external training dataset. We implemented CoCoA for widefield
imaging of mouse brain tissues and validated its performance with
direct-wavefront-sensing-based adaptive optics. Importantly, we systematically
explored and quantitatively characterized the limiting factors of CoCoA's
performance. Using CoCoA, we demonstrated the first in vivo widefield mouse
brain imaging using machine-learning-based adaptive optics. Incorporating
coordinate-based neural representations and a forward physics model, the
self-supervised scheme of CoCoA should be applicable to microscopy modalities
in general.
| [
{
"version": "v1",
"created": "Fri, 7 Jul 2023 19:36:24 GMT"
},
{
"version": "v2",
"created": "Thu, 25 Apr 2024 04:49:04 GMT"
},
{
"version": "v3",
"created": "Wed, 1 May 2024 23:31:41 GMT"
},
{
"version": "v4",
"created": "Mon, 24 Jun 2024 23:08:29 GMT"
},
{
"version": "v5",
"created": "Fri, 7 Mar 2025 20:29:51 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Kang",
"Iksung",
""
],
[
"Zhang",
"Qinrong",
""
],
[
"Yu",
"Stella X.",
""
],
[
"Ji",
"Na",
""
]
]
| TITLE: Coordinate-based neural representations for computational adaptive
optics in widefield microscopy
ABSTRACT: Widefield microscopy is widely used for non-invasive imaging of biological
structures at subcellular resolution. When applied to complex specimen, its
image quality is degraded by sample-induced optical aberration. Adaptive optics
can correct wavefront distortion and restore diffraction-limited resolution but
require wavefront sensing and corrective devices, increasing system complexity
and cost. Here, we describe a self-supervised machine learning algorithm,
CoCoA, that performs joint wavefront estimation and three-dimensional
structural information extraction from a single input 3D image stack without
the need for external training dataset. We implemented CoCoA for widefield
imaging of mouse brain tissues and validated its performance with
direct-wavefront-sensing-based adaptive optics. Importantly, we systematically
explored and quantitatively characterized the limiting factors of CoCoA's
performance. Using CoCoA, we demonstrated the first in vivo widefield mouse
brain imaging using machine-learning-based adaptive optics. Incorporating
coordinate-based neural representations and a forward physics model, the
self-supervised scheme of CoCoA should be applicable to microscopy modalities
in general.
| no_new_dataset | 0.948489 |
2310.08051 | Yuzhe Tian | Jianchao Lu and Yuzhe Tian, Yang Zhang, Quan Z. Sheng, Xi Zheng | LGL-BCI: A Motor-Imagery-Based Brain-Computer Interface with Geometric
Learning | Update the venue and copyright information | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Brain--computer interfaces are groundbreaking technology whereby brain
signals are used to control external devices. Despite some advances in recent
years, electroencephalogram (EEG)-based motor-imagery tasks face challenges,
such as amplitude and phase variability and complex spatial correlations, with
a need for smaller models and faster inference. In this study, we develop a
prototype, called the Lightweight Geometric Learning Brain--Computer Interface
(LGL-BCI), which uses our customized geometric deep learning architecture for
swift model inference without sacrificing accuracy. LGL-BCI contains an EEG
channel selection module via a feature decomposition algorithm to reduce the
dimensionality of a symmetric positive definite matrix, providing adaptiveness
among the continuously changing EEG signal. Meanwhile, a built-in lossless
transformation helps boost the inference speed. The performance of our solution
was evaluated using two real-world EEG devices and two public EEG datasets.
LGL-BCI demonstrated significant improvements, achieving an accuracy of 82.54%
compared to 62.22% for the state-of-the-art approach. Furthermore, LGL-BCI uses
fewer parameters (64.9K vs. 183.7K), highlighting its computational efficiency.
These findings underscore both the superior accuracy and computational
efficiency of LGL-BCI, demonstrating the feasibility and robustness of
geometric deep learning in motor-imagery brain--computer interface
applications.
| [
{
"version": "v1",
"created": "Thu, 12 Oct 2023 05:52:54 GMT"
},
{
"version": "v2",
"created": "Wed, 8 Nov 2023 05:30:25 GMT"
},
{
"version": "v3",
"created": "Tue, 21 Nov 2023 12:36:49 GMT"
},
{
"version": "v4",
"created": "Wed, 26 Feb 2025 05:16:11 GMT"
},
{
"version": "v5",
"created": "Sat, 8 Mar 2025 15:14:27 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Lu",
"Jianchao",
""
],
[
"Tian",
"Yuzhe",
""
],
[
"Zhang",
"Yang",
""
],
[
"Sheng",
"Quan Z.",
""
],
[
"Zheng",
"Xi",
""
]
]
| TITLE: LGL-BCI: A Motor-Imagery-Based Brain-Computer Interface with Geometric
Learning
ABSTRACT: Brain--computer interfaces are groundbreaking technology whereby brain
signals are used to control external devices. Despite some advances in recent
years, electroencephalogram (EEG)-based motor-imagery tasks face challenges,
such as amplitude and phase variability and complex spatial correlations, with
a need for smaller models and faster inference. In this study, we develop a
prototype, called the Lightweight Geometric Learning Brain--Computer Interface
(LGL-BCI), which uses our customized geometric deep learning architecture for
swift model inference without sacrificing accuracy. LGL-BCI contains an EEG
channel selection module via a feature decomposition algorithm to reduce the
dimensionality of a symmetric positive definite matrix, providing adaptiveness
among the continuously changing EEG signal. Meanwhile, a built-in lossless
transformation helps boost the inference speed. The performance of our solution
was evaluated using two real-world EEG devices and two public EEG datasets.
LGL-BCI demonstrated significant improvements, achieving an accuracy of 82.54%
compared to 62.22% for the state-of-the-art approach. Furthermore, LGL-BCI uses
fewer parameters (64.9K vs. 183.7K), highlighting its computational efficiency.
These findings underscore both the superior accuracy and computational
efficiency of LGL-BCI, demonstrating the feasibility and robustness of
geometric deep learning in motor-imagery brain--computer interface
applications.
| no_new_dataset | 0.949201 |
2310.12214 | Meng Tong | Meng Tong and Kejiang Chen and Jie Zhang and Yuang Qi and Weiming
Zhang and Nenghai Yu and Tianwei Zhang and Zhikun Zhang | InferDPT: Privacy-Preserving Inference for Black-box Large Language
Model | null | null | null | null | cs.CR | http://creativecommons.org/licenses/by/4.0/ | Large language models (LLMs), like ChatGPT, have greatly simplified text
generation tasks. However, they have also raised concerns about privacy risks
such as data leakage and unauthorized data collection. Existing solutions for
privacy-preserving inference face practical challenges related to computation
time and communication costs. In this paper, we propose InferDPT, the first
practical framework for the privacy-preserving Inference of black-box LLMs,
implementing Differential Privacy in Text generation. InferDPT comprises two
key modules: the "perturbation module" utilizes the exponential mechanism to
generate a perturbed prompt, facilitating privacy-preserving inference with
black-box LLMs, and the "extraction module", inspired by knowledge distillation
and retrieval-augmented generation, extracts coherent and consistent text from
the perturbed generation result, ensuring successful text generation
completion. To address privacy concerns related to previous exponential
mechanisms' susceptibility to embedding revision attacks, we introduce RANTEXT,
a novel differential privacy mechanism integrated into the perturbation module
of InferDPT, which introduces the concept of "RANdom adjacency" for TEXT
perturbation within the prompt. Experimental results across three datasets
demonstrate that the text generation quality of InferDPT is comparable to that
of non-private GPT-4, and RANTEXT surpasses existing state-of-the-art
mechanisms, namely, SANTEXT+ and CUSTEXT+ in the trade-off between privacy and
utility. Even with an privacy parameter epsilon value of 6.0, RANTEXT achieves
an average privacy protection rate exceeding 90% against embedding revision
attacks, which is 0.58 times higher than that of SANTEXT+ and 3.35 times higher
than that of CUSTEXT+.
| [
{
"version": "v1",
"created": "Wed, 18 Oct 2023 18:00:11 GMT"
},
{
"version": "v2",
"created": "Sun, 22 Oct 2023 07:34:36 GMT"
},
{
"version": "v3",
"created": "Tue, 24 Oct 2023 03:25:14 GMT"
},
{
"version": "v4",
"created": "Fri, 8 Dec 2023 05:14:40 GMT"
},
{
"version": "v5",
"created": "Mon, 11 Dec 2023 09:59:09 GMT"
},
{
"version": "v6",
"created": "Wed, 27 Mar 2024 09:19:01 GMT"
},
{
"version": "v7",
"created": "Mon, 10 Mar 2025 06:52:58 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Tong",
"Meng",
""
],
[
"Chen",
"Kejiang",
""
],
[
"Zhang",
"Jie",
""
],
[
"Qi",
"Yuang",
""
],
[
"Zhang",
"Weiming",
""
],
[
"Yu",
"Nenghai",
""
],
[
"Zhang",
"Tianwei",
""
],
[
"Zhang",
"Zhikun",
""
]
]
| TITLE: InferDPT: Privacy-Preserving Inference for Black-box Large Language
Model
ABSTRACT: Large language models (LLMs), like ChatGPT, have greatly simplified text
generation tasks. However, they have also raised concerns about privacy risks
such as data leakage and unauthorized data collection. Existing solutions for
privacy-preserving inference face practical challenges related to computation
time and communication costs. In this paper, we propose InferDPT, the first
practical framework for the privacy-preserving Inference of black-box LLMs,
implementing Differential Privacy in Text generation. InferDPT comprises two
key modules: the "perturbation module" utilizes the exponential mechanism to
generate a perturbed prompt, facilitating privacy-preserving inference with
black-box LLMs, and the "extraction module", inspired by knowledge distillation
and retrieval-augmented generation, extracts coherent and consistent text from
the perturbed generation result, ensuring successful text generation
completion. To address privacy concerns related to previous exponential
mechanisms' susceptibility to embedding revision attacks, we introduce RANTEXT,
a novel differential privacy mechanism integrated into the perturbation module
of InferDPT, which introduces the concept of "RANdom adjacency" for TEXT
perturbation within the prompt. Experimental results across three datasets
demonstrate that the text generation quality of InferDPT is comparable to that
of non-private GPT-4, and RANTEXT surpasses existing state-of-the-art
mechanisms, namely, SANTEXT+ and CUSTEXT+ in the trade-off between privacy and
utility. Even with an privacy parameter epsilon value of 6.0, RANTEXT achieves
an average privacy protection rate exceeding 90% against embedding revision
attacks, which is 0.58 times higher than that of SANTEXT+ and 3.35 times higher
than that of CUSTEXT+.
| no_new_dataset | 0.948632 |
2311.17093 | Evelyn Mannix | Evelyn Mannix and Howard Bondell | A Mixture of Exemplars Approach for Efficient Out-of-Distribution
Detection with Foundation Models | null | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | One of the early weaknesses identified in deep neural networks trained for
image classification tasks was their inability to provide low confidence
predictions on out-of-distribution (OOD) data that was significantly different
from the in-distribution (ID) data used to train them. Representation learning,
where neural networks are trained in specific ways that improve their ability
to detect OOD examples, has emerged as a promising solution. However, these
approaches require long training times and can add additional overhead to
detect OOD examples. Recent developments in Vision Transformer (ViT) foundation
models$\unicode{x2013}$large networks trained on large and diverse datasets
with self-supervised approaches$\unicode{x2013}$also show strong performance in
OOD detection, and could address these challenges. This paper presents Mixture
of Exemplars (MoLAR), an efficient approach to tackling OOD detection
challenges that is designed to maximise the benefit of training a classifier
with a high quality, frozen, pretrained foundation model backbone. MoLAR
provides strong OOD performance when only comparing the similarity of OOD
examples to the exemplars, a small set of images chosen to be representative of
the dataset, leading to up to 30 times faster OOD detection inference over
other methods that provide best performance when the full ID dataset is used.
In some cases, only using these exemplars actually improves performance with
MoLAR. Extensive experiments demonstrate the improved OOD detection performance
of MoLAR in comparison to comparable approaches in both supervised and
semi-supervised settings, and code is available at
github.com/emannix/molar-mixture-of-exemplars.
| [
{
"version": "v1",
"created": "Tue, 28 Nov 2023 06:12:28 GMT"
},
{
"version": "v2",
"created": "Thu, 7 Mar 2024 00:30:52 GMT"
},
{
"version": "v3",
"created": "Fri, 24 May 2024 06:06:34 GMT"
},
{
"version": "v4",
"created": "Fri, 22 Nov 2024 01:20:29 GMT"
},
{
"version": "v5",
"created": "Sat, 8 Mar 2025 00:58:33 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Mannix",
"Evelyn",
""
],
[
"Bondell",
"Howard",
""
]
]
| TITLE: A Mixture of Exemplars Approach for Efficient Out-of-Distribution
Detection with Foundation Models
ABSTRACT: One of the early weaknesses identified in deep neural networks trained for
image classification tasks was their inability to provide low confidence
predictions on out-of-distribution (OOD) data that was significantly different
from the in-distribution (ID) data used to train them. Representation learning,
where neural networks are trained in specific ways that improve their ability
to detect OOD examples, has emerged as a promising solution. However, these
approaches require long training times and can add additional overhead to
detect OOD examples. Recent developments in Vision Transformer (ViT) foundation
models$\unicode{x2013}$large networks trained on large and diverse datasets
with self-supervised approaches$\unicode{x2013}$also show strong performance in
OOD detection, and could address these challenges. This paper presents Mixture
of Exemplars (MoLAR), an efficient approach to tackling OOD detection
challenges that is designed to maximise the benefit of training a classifier
with a high quality, frozen, pretrained foundation model backbone. MoLAR
provides strong OOD performance when only comparing the similarity of OOD
examples to the exemplars, a small set of images chosen to be representative of
the dataset, leading to up to 30 times faster OOD detection inference over
other methods that provide best performance when the full ID dataset is used.
In some cases, only using these exemplars actually improves performance with
MoLAR. Extensive experiments demonstrate the improved OOD detection performance
of MoLAR in comparison to comparable approaches in both supervised and
semi-supervised settings, and code is available at
github.com/emannix/molar-mixture-of-exemplars.
| no_new_dataset | 0.948822 |
2311.17750 | Gergely D\'aniel N\'emeth | Gergely D\'aniel N\'emeth, Miguel \'Angel Lozano, Novi Quadrianto,
Nuria Oliver | Privacy and Accuracy Implications of Model Complexity and Integration in
Heterogeneous Federated Learning | Code: https://github.com/ellisalicante/ma-fl-mia | IEEE Access 13 (2025) 40258-40274 | 10.1109/ACCESS.2025.3546478 | null | cs.LG cs.AI cs.CR | http://creativecommons.org/licenses/by/4.0/ | Federated Learning (FL) has been proposed as a privacy-preserving solution
for distributed machine learning, particularly in heterogeneous FL settings
where clients have varying computational capabilities and thus train models
with different complexities compared to the server's model. However, FL is not
without vulnerabilities: recent studies have shown that it is susceptible to
membership inference attacks (MIA), which can compromise the privacy of client
data. In this paper, we examine the intersection of these two aspects,
heterogeneous FL and its privacy vulnerabilities, by focusing on the role of
client model integration, the process through which the server integrates
parameters from clients' smaller models into its larger model. To better
understand this process, we first propose a taxonomy that categorizes existing
heterogeneous FL methods and enables the design of seven novel heterogeneous FL
model integration strategies. Using CIFAR-10, CIFAR-100, and FEMNIST vision
datasets, we evaluate the privacy and accuracy trade-offs of these approaches
under three types of MIAs. Our findings reveal significant differences in
privacy leakage and performance depending on the integration method. Notably,
introducing randomness in the model integration process enhances client privacy
while maintaining competitive accuracy for both the clients and the server.
This work provides quantitative light on the privacy-accuracy implications
client model integration in heterogeneous FL settings, paving the way towards
more secure and efficient FL systems.
| [
{
"version": "v1",
"created": "Wed, 29 Nov 2023 15:54:15 GMT"
},
{
"version": "v2",
"created": "Thu, 4 Jul 2024 08:33:33 GMT"
},
{
"version": "v3",
"created": "Mon, 10 Mar 2025 11:10:50 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Németh",
"Gergely Dániel",
""
],
[
"Lozano",
"Miguel Ángel",
""
],
[
"Quadrianto",
"Novi",
""
],
[
"Oliver",
"Nuria",
""
]
]
| TITLE: Privacy and Accuracy Implications of Model Complexity and Integration in
Heterogeneous Federated Learning
ABSTRACT: Federated Learning (FL) has been proposed as a privacy-preserving solution
for distributed machine learning, particularly in heterogeneous FL settings
where clients have varying computational capabilities and thus train models
with different complexities compared to the server's model. However, FL is not
without vulnerabilities: recent studies have shown that it is susceptible to
membership inference attacks (MIA), which can compromise the privacy of client
data. In this paper, we examine the intersection of these two aspects,
heterogeneous FL and its privacy vulnerabilities, by focusing on the role of
client model integration, the process through which the server integrates
parameters from clients' smaller models into its larger model. To better
understand this process, we first propose a taxonomy that categorizes existing
heterogeneous FL methods and enables the design of seven novel heterogeneous FL
model integration strategies. Using CIFAR-10, CIFAR-100, and FEMNIST vision
datasets, we evaluate the privacy and accuracy trade-offs of these approaches
under three types of MIAs. Our findings reveal significant differences in
privacy leakage and performance depending on the integration method. Notably,
introducing randomness in the model integration process enhances client privacy
while maintaining competitive accuracy for both the clients and the server.
This work provides quantitative light on the privacy-accuracy implications
client model integration in heterogeneous FL settings, paving the way towards
more secure and efficient FL systems.
| no_new_dataset | 0.945147 |
2312.05657 | Nikos Kanakaris | Shukai Duan, Nikos Kanakaris, Xiongye Xiao, Heng Ping, Chenyu Zhou,
Nesreen K. Ahmed, Guixiang Ma, Mihai Capota, Theodore L. Willke, Shahin
Nazarian, Paul Bogdan | PerfRL: A Small Language Model Framework for Efficient Code Optimization | null | null | null | null | cs.LG cs.AI cs.PL cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Code optimization is a challenging task requiring a substantial level of
expertise from developers. Nonetheless, this level of human capacity is not
sufficient considering the rapid evolution of new hardware architectures and
software environments. In light of this, recent research proposes adopting
machine learning and artificial intelligence techniques to automate the code
optimization process. In this paper, we introduce PerfRL, an innovative
framework designed to tackle the problem of code optimization. Our framework
leverages the capabilities of small language models (SLMs) and reinforcement
learning (RL), facilitating a system where SLMs can assimilate feedback from
their environment during the fine-tuning phase, notably through unit tests.
When benchmarked against existing models, PerfRL demonstrates superior
efficiency in terms of speed and computational resource usage, attributed to
its reduced need for training steps and its compatibility with SLMs.
Furthermore, it substantially diminishes the risk of logical and syntactical
errors. To evaluate our framework, we conduct experiments on the PIE dataset
using a lightweight large language model (i.e., CodeT5) and a new reinforcement
learning algorithm, namely RRHF. For evaluation purposes, we use a list of
evaluation metrics related to optimization quality and speedup. The evaluation
results show that our approach achieves similar or better results compared to
state-of-the-art models using shorter training times and smaller pre-trained
models.
| [
{
"version": "v1",
"created": "Sat, 9 Dec 2023 19:50:23 GMT"
},
{
"version": "v2",
"created": "Sun, 9 Mar 2025 05:01:42 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Duan",
"Shukai",
""
],
[
"Kanakaris",
"Nikos",
""
],
[
"Xiao",
"Xiongye",
""
],
[
"Ping",
"Heng",
""
],
[
"Zhou",
"Chenyu",
""
],
[
"Ahmed",
"Nesreen K.",
""
],
[
"Ma",
"Guixiang",
""
],
[
"Capota",
"Mihai",
""
],
[
"Willke",
"Theodore L.",
""
],
[
"Nazarian",
"Shahin",
""
],
[
"Bogdan",
"Paul",
""
]
]
| TITLE: PerfRL: A Small Language Model Framework for Efficient Code Optimization
ABSTRACT: Code optimization is a challenging task requiring a substantial level of
expertise from developers. Nonetheless, this level of human capacity is not
sufficient considering the rapid evolution of new hardware architectures and
software environments. In light of this, recent research proposes adopting
machine learning and artificial intelligence techniques to automate the code
optimization process. In this paper, we introduce PerfRL, an innovative
framework designed to tackle the problem of code optimization. Our framework
leverages the capabilities of small language models (SLMs) and reinforcement
learning (RL), facilitating a system where SLMs can assimilate feedback from
their environment during the fine-tuning phase, notably through unit tests.
When benchmarked against existing models, PerfRL demonstrates superior
efficiency in terms of speed and computational resource usage, attributed to
its reduced need for training steps and its compatibility with SLMs.
Furthermore, it substantially diminishes the risk of logical and syntactical
errors. To evaluate our framework, we conduct experiments on the PIE dataset
using a lightweight large language model (i.e., CodeT5) and a new reinforcement
learning algorithm, namely RRHF. For evaluation purposes, we use a list of
evaluation metrics related to optimization quality and speedup. The evaluation
results show that our approach achieves similar or better results compared to
state-of-the-art models using shorter training times and smaller pre-trained
models.
| no_new_dataset | 0.942612 |
2312.13440 | Tonmoy Hossain | Tonmoy Hossain and Miaomiao Zhang | MGAug: Multimodal Geometric Augmentation in Latent Spaces of Image
Deformations | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Geometric transformations have been widely used to augment the size of
training images. Existing methods often assume a unimodal distribution of the
underlying transformations between images, which limits their power when data
with multimodal distributions occur. In this paper, we propose a novel model,
Multimodal Geometric Augmentation (MGAug), that for the first time generates
augmenting transformations in a multimodal latent space of geometric
deformations. To achieve this, we first develop a deep network that embeds the
learning of latent geometric spaces of diffeomorphic transformations (a.k.a.
diffeomorphisms) in a variational autoencoder (VAE). A mixture of multivariate
Gaussians is formulated in the tangent space of diffeomorphisms and serves as a
prior to approximate the hidden distribution of image transformations. We then
augment the original training dataset by deforming images using randomly
sampled transformations from the learned multimodal latent space of VAE. To
validate the efficiency of our model, we jointly learn the augmentation
strategy with two distinct domain-specific tasks: multi-class classification on
2D synthetic datasets and segmentation on real 3D brain magnetic resonance
images (MRIs). We also compare MGAug with state-of-the-art transformation-based
image augmentation algorithms. Experimental results show that our proposed
approach outperforms all baselines by significantly improved prediction
accuracy. Our code is publicly available at
https://github.com/tonmoy-hossain/MGAug.
| [
{
"version": "v1",
"created": "Wed, 20 Dec 2023 21:30:55 GMT"
},
{
"version": "v2",
"created": "Thu, 25 Jan 2024 18:31:49 GMT"
},
{
"version": "v3",
"created": "Sun, 9 Mar 2025 07:55:41 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Hossain",
"Tonmoy",
""
],
[
"Zhang",
"Miaomiao",
""
]
]
| TITLE: MGAug: Multimodal Geometric Augmentation in Latent Spaces of Image
Deformations
ABSTRACT: Geometric transformations have been widely used to augment the size of
training images. Existing methods often assume a unimodal distribution of the
underlying transformations between images, which limits their power when data
with multimodal distributions occur. In this paper, we propose a novel model,
Multimodal Geometric Augmentation (MGAug), that for the first time generates
augmenting transformations in a multimodal latent space of geometric
deformations. To achieve this, we first develop a deep network that embeds the
learning of latent geometric spaces of diffeomorphic transformations (a.k.a.
diffeomorphisms) in a variational autoencoder (VAE). A mixture of multivariate
Gaussians is formulated in the tangent space of diffeomorphisms and serves as a
prior to approximate the hidden distribution of image transformations. We then
augment the original training dataset by deforming images using randomly
sampled transformations from the learned multimodal latent space of VAE. To
validate the efficiency of our model, we jointly learn the augmentation
strategy with two distinct domain-specific tasks: multi-class classification on
2D synthetic datasets and segmentation on real 3D brain magnetic resonance
images (MRIs). We also compare MGAug with state-of-the-art transformation-based
image augmentation algorithms. Experimental results show that our proposed
approach outperforms all baselines by significantly improved prediction
accuracy. Our code is publicly available at
https://github.com/tonmoy-hossain/MGAug.
| no_new_dataset | 0.948775 |
2312.15497 | Corneliu Arsene Dr | Corneliu Arsene, Alessandra Parisio | Deep Convolutional Neural Networks for Short-Term Multi-Energy Demand
Prediction of Integrated Energy Systems | 29 pages, 40 figures | null | null | null | cs.LG cs.CE | http://creativecommons.org/licenses/by/4.0/ | Forecasting power consumptions of integrated electrical, heat or gas network
systems is essential in order to operate more efficiently the whole energy
network. Multi-energy systems are increasingly seen as a key component of
future energy systems, and a valuable source of flexibility, which can
significantly contribute to a cleaner and more sustainable whole energy system.
Therefore, there is a stringent need for developing novel and performant models
for forecasting multi-energy demand of integrated energy systems, which to
account for the different types of interacting energy vectors and of the
coupling between them. Previous efforts in demand forecasting focused mainly on
the single electrical power consumption or, more recently, on the single heat
or gas power consumptions. In order to address this gap, in this paper six
novel prediction models based on Convolutional Neural Networks (CNNs) are
developed, for either individual or joint prediction of multi-energy power
consumptions: the single input/single output CNN model with determining the
optimum number of epochs (CNN_1), the multiple input/single output CNN model
(CNN_2), the single input/ single output CNN model with
training/validation/testing datasets (CNN_3), the joint prediction CNN model
(CNN_4), the multiple-building input/output CNN model (CNN_5) and the federated
learning CNN model (CNN_6). All six novel CNN models are applied in a
comprehensive manner on a novel integrated electrical, heat and gas network
system, which only recently has started to be used for forecasting. The
forecast horizon is short-term (next half an hour) and all the predictions
results are evaluated in terms of the Signal to Noise Ratio (SNR) and the
Normalized Root Mean Square Error (NRMSE), while the Mean Absolute Percentage
Error (MAPE) is used for comparison purposes with other existent results from
literature.
| [
{
"version": "v1",
"created": "Sun, 24 Dec 2023 14:56:23 GMT"
},
{
"version": "v2",
"created": "Mon, 10 Mar 2025 14:05:46 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Arsene",
"Corneliu",
""
],
[
"Parisio",
"Alessandra",
""
]
]
| TITLE: Deep Convolutional Neural Networks for Short-Term Multi-Energy Demand
Prediction of Integrated Energy Systems
ABSTRACT: Forecasting power consumptions of integrated electrical, heat or gas network
systems is essential in order to operate more efficiently the whole energy
network. Multi-energy systems are increasingly seen as a key component of
future energy systems, and a valuable source of flexibility, which can
significantly contribute to a cleaner and more sustainable whole energy system.
Therefore, there is a stringent need for developing novel and performant models
for forecasting multi-energy demand of integrated energy systems, which to
account for the different types of interacting energy vectors and of the
coupling between them. Previous efforts in demand forecasting focused mainly on
the single electrical power consumption or, more recently, on the single heat
or gas power consumptions. In order to address this gap, in this paper six
novel prediction models based on Convolutional Neural Networks (CNNs) are
developed, for either individual or joint prediction of multi-energy power
consumptions: the single input/single output CNN model with determining the
optimum number of epochs (CNN_1), the multiple input/single output CNN model
(CNN_2), the single input/ single output CNN model with
training/validation/testing datasets (CNN_3), the joint prediction CNN model
(CNN_4), the multiple-building input/output CNN model (CNN_5) and the federated
learning CNN model (CNN_6). All six novel CNN models are applied in a
comprehensive manner on a novel integrated electrical, heat and gas network
system, which only recently has started to be used for forecasting. The
forecast horizon is short-term (next half an hour) and all the predictions
results are evaluated in terms of the Signal to Noise Ratio (SNR) and the
Normalized Root Mean Square Error (NRMSE), while the Mean Absolute Percentage
Error (MAPE) is used for comparison purposes with other existent results from
literature.
| no_new_dataset | 0.952662 |
2312.16810 | Hanqi Su | Hanqi Su, Jay Lee | Machine Learning Approaches for Diagnostics and Prognostics of
Industrial Systems Using Open Source Data from PHM Data Challenges: A Review | The paper submitted to the International Journal of Prognostics and
Health Management (IJPHM) has been accepted | International Journal of Prognostics and Health Management, Volume
15, Issue 2, 2024 | 10.36001/ijphm.2024.v15i2.3993 | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | In the field of Prognostics and Health Management (PHM), recent years have
witnessed a significant surge in the application of machine learning (ML).
Despite this growth, the field grapples with a lack of unified guidelines and
systematic approaches for effectively implementing these ML techniques and
comprehensive analysis regarding industrial open-source data across varied
scenarios. To address these gaps, this paper provides a comprehensive review of
ML approaches for diagnostics and prognostics of industrial systems using
open-source datasets from PHM Data Challenge Competitions held between 2018 and
2023 by PHM Society and IEEE Reliability Society and summarizes a unified ML
framework. This review systematically categorizes and scrutinizes the problems,
challenges, methodologies, and advancements demonstrated in these competitions,
highlighting the evolving role of both conventional machine learning and deep
learning in tackling complex industrial tasks related to detection, diagnosis,
assessment, and prognosis. Moreover, this paper delves into the common
challenges in PHM data challenge competitions by emphasizing data-related and
model-related issues and evaluating the limitations of these competitions. The
potential solutions to address these challenges are also summarized. Finally,
we identify key themes and potential directions for future research, providing
opportunities and prospects for next-generation ML-PHM development in PHM
domain.
| [
{
"version": "v1",
"created": "Thu, 28 Dec 2023 04:00:25 GMT"
},
{
"version": "v2",
"created": "Fri, 24 May 2024 21:09:34 GMT"
},
{
"version": "v3",
"created": "Wed, 18 Sep 2024 17:45:20 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Su",
"Hanqi",
""
],
[
"Lee",
"Jay",
""
]
]
| TITLE: Machine Learning Approaches for Diagnostics and Prognostics of
Industrial Systems Using Open Source Data from PHM Data Challenges: A Review
ABSTRACT: In the field of Prognostics and Health Management (PHM), recent years have
witnessed a significant surge in the application of machine learning (ML).
Despite this growth, the field grapples with a lack of unified guidelines and
systematic approaches for effectively implementing these ML techniques and
comprehensive analysis regarding industrial open-source data across varied
scenarios. To address these gaps, this paper provides a comprehensive review of
ML approaches for diagnostics and prognostics of industrial systems using
open-source datasets from PHM Data Challenge Competitions held between 2018 and
2023 by PHM Society and IEEE Reliability Society and summarizes a unified ML
framework. This review systematically categorizes and scrutinizes the problems,
challenges, methodologies, and advancements demonstrated in these competitions,
highlighting the evolving role of both conventional machine learning and deep
learning in tackling complex industrial tasks related to detection, diagnosis,
assessment, and prognosis. Moreover, this paper delves into the common
challenges in PHM data challenge competitions by emphasizing data-related and
model-related issues and evaluating the limitations of these competitions. The
potential solutions to address these challenges are also summarized. Finally,
we identify key themes and potential directions for future research, providing
opportunities and prospects for next-generation ML-PHM development in PHM
domain.
| no_new_dataset | 0.943971 |
2401.00260 | Hao Ruan | Jun Wang, Hao Ruan, Liangjian Wen, Yong Dai, Mingjie Wang | GazeCLIP: Enhancing Gaze Estimation Through Text-Guided Multimodal
Learning | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Visual gaze estimation, with its wide-ranging application scenarios, has
garnered increasing attention within the research community. Although existing
approaches infer gaze solely from image signals, recent advances in
visual-language collaboration have demonstrated that the integration of
linguistic information can significantly enhance performance across various
visual tasks. Leveraging the remarkable transferability of large-scale
Contrastive Language-Image Pre-training (CLIP) models, we address the open and
urgent question of how to effectively apply linguistic cues to gaze estimation.
In this work, we propose GazeCLIP, a novel gaze estimation framework that
deeply explores text-face collaboration. Specifically, we introduce a
meticulously designed linguistic description generator to produce text signals
enriched with coarse directional cues. Furthermore, we present a CLIP-based
backbone adept at characterizing text-face pairs for gaze estimation,
complemented by a fine-grained multimodal fusion module that models the
intricate interrelationships between heterogeneous inputs. Extensive
experiments on three challenging datasets demonstrate the superiority of
GazeCLIP, which achieves state-of-the-art accuracy. Our findings underscore the
potential of using visual-language collaboration to advance gaze estimation and
open new avenues for future research in multimodal learning for visual tasks.
The implementation code and the pre-trained model will be made publicly
available.
| [
{
"version": "v1",
"created": "Sat, 30 Dec 2023 15:24:50 GMT"
},
{
"version": "v2",
"created": "Sun, 7 Jan 2024 04:17:20 GMT"
},
{
"version": "v3",
"created": "Fri, 26 Apr 2024 03:59:41 GMT"
},
{
"version": "v4",
"created": "Sat, 8 Mar 2025 13:37:22 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Wang",
"Jun",
""
],
[
"Ruan",
"Hao",
""
],
[
"Wen",
"Liangjian",
""
],
[
"Dai",
"Yong",
""
],
[
"Wang",
"Mingjie",
""
]
]
| TITLE: GazeCLIP: Enhancing Gaze Estimation Through Text-Guided Multimodal
Learning
ABSTRACT: Visual gaze estimation, with its wide-ranging application scenarios, has
garnered increasing attention within the research community. Although existing
approaches infer gaze solely from image signals, recent advances in
visual-language collaboration have demonstrated that the integration of
linguistic information can significantly enhance performance across various
visual tasks. Leveraging the remarkable transferability of large-scale
Contrastive Language-Image Pre-training (CLIP) models, we address the open and
urgent question of how to effectively apply linguistic cues to gaze estimation.
In this work, we propose GazeCLIP, a novel gaze estimation framework that
deeply explores text-face collaboration. Specifically, we introduce a
meticulously designed linguistic description generator to produce text signals
enriched with coarse directional cues. Furthermore, we present a CLIP-based
backbone adept at characterizing text-face pairs for gaze estimation,
complemented by a fine-grained multimodal fusion module that models the
intricate interrelationships between heterogeneous inputs. Extensive
experiments on three challenging datasets demonstrate the superiority of
GazeCLIP, which achieves state-of-the-art accuracy. Our findings underscore the
potential of using visual-language collaboration to advance gaze estimation and
open new avenues for future research in multimodal learning for visual tasks.
The implementation code and the pre-trained model will be made publicly
available.
| no_new_dataset | 0.937669 |
2401.10893 | Deepak Banerjee | Deepak Banerjee, Anjali Ishaan | Location Sensitive Embedding for Knowledge Graph Reasoning | null | null | null | null | cs.IR cs.CL | http://creativecommons.org/licenses/by/4.0/ | Embedding methods transform the knowledge graph into a continuous,
low-dimensional space, facilitating inference and completion tasks. Existing
methods are mainly divided into two types: translational distance models and
semantic matching models. A key challenge in translational distance models is
their inability to effectively differentiate between 'head' and 'tail' entities
in graphs. To address this problem, a novel location-sensitive embedding (LSE)
method has been developed. LSE innovatively modifies the head entity using
relation-specific mappings, conceptualizing relations as linear transformations
rather than mere translations. The theoretical foundations of LSE, including
its representational capabilities and its connections to existing models, have
been thoroughly examined. A more streamlined variant, LSEd, which employs a
diagonal matrix for transformations to enhance practical efficiency, is also
proposed. Experiments conducted on four large-scale KG datasets for link
prediction show that LSEd either outperforms or is competitive with
state-of-the-art related works.
| [
{
"version": "v1",
"created": "Fri, 1 Dec 2023 22:35:19 GMT"
},
{
"version": "v2",
"created": "Sat, 27 Jan 2024 22:25:09 GMT"
},
{
"version": "v3",
"created": "Tue, 30 Jan 2024 03:14:11 GMT"
},
{
"version": "v4",
"created": "Sat, 8 Mar 2025 00:43:01 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Banerjee",
"Deepak",
""
],
[
"Ishaan",
"Anjali",
""
]
]
| TITLE: Location Sensitive Embedding for Knowledge Graph Reasoning
ABSTRACT: Embedding methods transform the knowledge graph into a continuous,
low-dimensional space, facilitating inference and completion tasks. Existing
methods are mainly divided into two types: translational distance models and
semantic matching models. A key challenge in translational distance models is
their inability to effectively differentiate between 'head' and 'tail' entities
in graphs. To address this problem, a novel location-sensitive embedding (LSE)
method has been developed. LSE innovatively modifies the head entity using
relation-specific mappings, conceptualizing relations as linear transformations
rather than mere translations. The theoretical foundations of LSE, including
its representational capabilities and its connections to existing models, have
been thoroughly examined. A more streamlined variant, LSEd, which employs a
diagonal matrix for transformations to enhance practical efficiency, is also
proposed. Experiments conducted on four large-scale KG datasets for link
prediction show that LSEd either outperforms or is competitive with
state-of-the-art related works.
| no_new_dataset | 0.943138 |
2401.15199 | Zahra Kharazian | Zahra Kharazian, Tony Lindgren, Sindri Magn\'usson, Olof Steinert,
Oskar Andersson Reyna | SCANIA Component X Dataset: A Real-World Multivariate Time Series
Dataset for Predictive Maintenance | 12 pages, 8 figures | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Predicting failures and maintenance time in predictive maintenance is
challenging due to the scarcity of comprehensive real-world datasets, and among
those available, few are of time series format. This paper introduces a
real-world, multivariate time series dataset collected exclusively from a
single anonymized engine component (Component X) across a fleet of SCANIA
trucks. The dataset includes operational data, repair records, and
specifications related to Component X, while maintaining confidentiality
through anonymization. It is well-suited for a range of machine learning
applications, including classification, regression, survival analysis, and
anomaly detection, particularly in predictive maintenance scenarios. The
dataset's large population size, diverse features (in the form of histograms
and numerical counters), and temporal information make it a unique resource in
the field. The objective of releasing this dataset is to give a broad range of
researchers the possibility of working with real-world data from an
internationally well-known company and introduce a standard benchmark to the
predictive maintenance field, fostering reproducible research.
| [
{
"version": "v1",
"created": "Fri, 26 Jan 2024 20:51:55 GMT"
},
{
"version": "v2",
"created": "Mon, 10 Mar 2025 09:12:04 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Kharazian",
"Zahra",
""
],
[
"Lindgren",
"Tony",
""
],
[
"Magnússon",
"Sindri",
""
],
[
"Steinert",
"Olof",
""
],
[
"Reyna",
"Oskar Andersson",
""
]
]
| TITLE: SCANIA Component X Dataset: A Real-World Multivariate Time Series
Dataset for Predictive Maintenance
ABSTRACT: Predicting failures and maintenance time in predictive maintenance is
challenging due to the scarcity of comprehensive real-world datasets, and among
those available, few are of time series format. This paper introduces a
real-world, multivariate time series dataset collected exclusively from a
single anonymized engine component (Component X) across a fleet of SCANIA
trucks. The dataset includes operational data, repair records, and
specifications related to Component X, while maintaining confidentiality
through anonymization. It is well-suited for a range of machine learning
applications, including classification, regression, survival analysis, and
anomaly detection, particularly in predictive maintenance scenarios. The
dataset's large population size, diverse features (in the form of histograms
and numerical counters), and temporal information make it a unique resource in
the field. The objective of releasing this dataset is to give a broad range of
researchers the possibility of working with real-world data from an
internationally well-known company and introduce a standard benchmark to the
predictive maintenance field, fostering reproducible research.
| new_dataset | 0.962532 |
2402.00906 | Hamed Poursiami | Hamed Poursiami, Ihsen Alouani, Maryam Parsa | BrainLeaks: On the Privacy-Preserving Properties of Neuromorphic
Architectures against Model Inversion Attacks | 7 pages, 4 figures, 4 tables | 2024 International Conference on Machine Learning and Applications
(ICMLA), 2024, pp. 705-712 | 10.1109/ICMLA61862.2024.00102 | null | cs.CR cs.LG cs.NE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | With the mainstream integration of machine learning into security-sensitive
domains such as healthcare and finance, concerns about data privacy have
intensified. Conventional artificial neural networks (ANNs) have been found
vulnerable to several attacks that can leak sensitive data. Particularly, model
inversion (MI) attacks enable the reconstruction of data samples that have been
used to train the model. Neuromorphic architectures have emerged as a paradigm
shift in neural computing, enabling asynchronous and energy-efficient
computation. However, little to no existing work has investigated the privacy
of neuromorphic architectures against model inversion. Our study is motivated
by the intuition that the non-differentiable aspect of spiking neural networks
(SNNs) might result in inherent privacy-preserving properties, especially
against gradient-based attacks. To investigate this hypothesis, we propose a
thorough exploration of SNNs' privacy-preserving capabilities. Specifically, we
develop novel inversion attack strategies that are comprehensively designed to
target SNNs, offering a comparative analysis with their conventional ANN
counterparts. Our experiments, conducted on diverse event-based and static
datasets, demonstrate the effectiveness of the proposed attack strategies and
therefore questions the assumption of inherent privacy-preserving in
neuromorphic architectures.
| [
{
"version": "v1",
"created": "Thu, 1 Feb 2024 03:16:40 GMT"
},
{
"version": "v2",
"created": "Tue, 7 May 2024 05:53:46 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Poursiami",
"Hamed",
""
],
[
"Alouani",
"Ihsen",
""
],
[
"Parsa",
"Maryam",
""
]
]
| TITLE: BrainLeaks: On the Privacy-Preserving Properties of Neuromorphic
Architectures against Model Inversion Attacks
ABSTRACT: With the mainstream integration of machine learning into security-sensitive
domains such as healthcare and finance, concerns about data privacy have
intensified. Conventional artificial neural networks (ANNs) have been found
vulnerable to several attacks that can leak sensitive data. Particularly, model
inversion (MI) attacks enable the reconstruction of data samples that have been
used to train the model. Neuromorphic architectures have emerged as a paradigm
shift in neural computing, enabling asynchronous and energy-efficient
computation. However, little to no existing work has investigated the privacy
of neuromorphic architectures against model inversion. Our study is motivated
by the intuition that the non-differentiable aspect of spiking neural networks
(SNNs) might result in inherent privacy-preserving properties, especially
against gradient-based attacks. To investigate this hypothesis, we propose a
thorough exploration of SNNs' privacy-preserving capabilities. Specifically, we
develop novel inversion attack strategies that are comprehensively designed to
target SNNs, offering a comparative analysis with their conventional ANN
counterparts. Our experiments, conducted on diverse event-based and static
datasets, demonstrate the effectiveness of the proposed attack strategies and
therefore questions the assumption of inherent privacy-preserving in
neuromorphic architectures.
| no_new_dataset | 0.947478 |
2402.07818 | Zhihao Liu | Z Liu, J Lou, W Bao, Y Hu, B Li, Z Qin, K Ren | Differentially Private Zeroth-Order Methods for Scalable Large Language
Model Finetuning | null | null | null | null | cs.LG cs.AI cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Fine-tuning on task-specific datasets is a widely-embraced paradigm of
harnessing the powerful capability of pretrained LLMs for various downstream
tasks. Due to the popularity of LLMs fine-tuning and its accompanying privacy
concerns, differentially private (DP) fine-tuning of pretrained LLMs has been
widely used to safeguarding the privacy of task-specific datasets. Lying at the
design core of DP LLM fine-tuning methods is the satisfactory tradeoff among
privacy, utility, and scalability. Most existing methods build upon the seminal
work of DP-SGD. Despite pushing the scalability of DP-SGD to its limit,
DP-SGD-based fine-tuning methods are unfortunately limited by the inherent
inefficiency of SGD.
In this paper, we investigate the potential of DP zeroth-order methods for
LLM pretraining, which avoids the scalability bottleneck of SGD by
approximating the gradient with the more efficient zeroth-order gradient.
Rather than treating the zeroth-order method as a drop-in replacement for SGD,
this paper presents a comprehensive study both theoretically and empirically.
First, we propose the stagewise DP zeroth-order method (DP-ZOSO) that
dynamically schedules key hyperparameters. This design is grounded on the
synergy between DP random perturbation and the gradient approximation error of
the zeroth-order method, and its effect on fine-tuning trajectory.
We provide theoretical analysis for both proposed methods. We conduct
extensive empirical analysis on both encoder-only masked language model and
decoder-only autoregressive language model, achieving impressive results in
terms of scalability and utility regardless of the class of tasks (compared
with DPZero, DP-ZOPO improves $4.5\%$ on SST-5, $5.5\%$ on MNLI with
RoBERTa-Large and 9.2\% on CB, 3.9\% on BoolQ with OPT-2.7b when $\epsilon=4$,
demonstrates more significant enhancement in performance on more complicated
tasks).
| [
{
"version": "v1",
"created": "Mon, 12 Feb 2024 17:24:15 GMT"
},
{
"version": "v2",
"created": "Wed, 21 Feb 2024 06:11:02 GMT"
},
{
"version": "v3",
"created": "Wed, 8 May 2024 07:14:42 GMT"
},
{
"version": "v4",
"created": "Thu, 9 May 2024 09:41:23 GMT"
},
{
"version": "v5",
"created": "Mon, 2 Dec 2024 12:29:47 GMT"
},
{
"version": "v6",
"created": "Mon, 10 Mar 2025 06:52:03 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Liu",
"Z",
""
],
[
"Lou",
"J",
""
],
[
"Bao",
"W",
""
],
[
"Hu",
"Y",
""
],
[
"Li",
"B",
""
],
[
"Qin",
"Z",
""
],
[
"Ren",
"K",
""
]
]
| TITLE: Differentially Private Zeroth-Order Methods for Scalable Large Language
Model Finetuning
ABSTRACT: Fine-tuning on task-specific datasets is a widely-embraced paradigm of
harnessing the powerful capability of pretrained LLMs for various downstream
tasks. Due to the popularity of LLMs fine-tuning and its accompanying privacy
concerns, differentially private (DP) fine-tuning of pretrained LLMs has been
widely used to safeguarding the privacy of task-specific datasets. Lying at the
design core of DP LLM fine-tuning methods is the satisfactory tradeoff among
privacy, utility, and scalability. Most existing methods build upon the seminal
work of DP-SGD. Despite pushing the scalability of DP-SGD to its limit,
DP-SGD-based fine-tuning methods are unfortunately limited by the inherent
inefficiency of SGD.
In this paper, we investigate the potential of DP zeroth-order methods for
LLM pretraining, which avoids the scalability bottleneck of SGD by
approximating the gradient with the more efficient zeroth-order gradient.
Rather than treating the zeroth-order method as a drop-in replacement for SGD,
this paper presents a comprehensive study both theoretically and empirically.
First, we propose the stagewise DP zeroth-order method (DP-ZOSO) that
dynamically schedules key hyperparameters. This design is grounded on the
synergy between DP random perturbation and the gradient approximation error of
the zeroth-order method, and its effect on fine-tuning trajectory.
We provide theoretical analysis for both proposed methods. We conduct
extensive empirical analysis on both encoder-only masked language model and
decoder-only autoregressive language model, achieving impressive results in
terms of scalability and utility regardless of the class of tasks (compared
with DPZero, DP-ZOPO improves $4.5\%$ on SST-5, $5.5\%$ on MNLI with
RoBERTa-Large and 9.2\% on CB, 3.9\% on BoolQ with OPT-2.7b when $\epsilon=4$,
demonstrates more significant enhancement in performance on more complicated
tasks).
| no_new_dataset | 0.951953 |
2402.09469 | Zhenmei Shi | Chenyang Li, Yingyu Liang, Zhenmei Shi, Zhao Song, Tianyi Zhou | Fourier Circuits in Neural Networks and Transformers: A Case Study of
Modular Arithmetic with Multiple Inputs | AIStats 2025 | null | null | null | cs.LG stat.ML | http://creativecommons.org/licenses/by/4.0/ | In the evolving landscape of machine learning, a pivotal challenge lies in
deciphering the internal representations harnessed by neural networks and
Transformers. Building on recent progress toward comprehending how networks
execute distinct target functions, our study embarks on an exploration of the
underlying reasons behind networks adopting specific computational strategies.
We direct our focus to the complex algebraic learning task of modular addition
involving $k$ inputs. Our research presents a thorough analytical
characterization of the features learned by stylized one-hidden layer neural
networks and one-layer Transformers in addressing this task. A cornerstone of
our theoretical framework is the elucidation of how the principle of margin
maximization shapes the features adopted by one-hidden layer neural networks.
Let $p$ denote the modulus, $D_p$ denote the dataset of modular arithmetic with
$k$ inputs and $m$ denote the network width. We demonstrate that a neuron count
of $ m \geq 2^{2k-2} \cdot (p-1) $, these networks attain a maximum $ L_{2,k+1}
$-margin on the dataset $ D_p $. Furthermore, we establish that each
hidden-layer neuron aligns with a specific Fourier spectrum, integral to
solving modular addition problems. By correlating our findings with the
empirical observations of similar studies, we contribute to a deeper
comprehension of the intrinsic computational mechanisms of neural networks.
Furthermore, we observe similar computational mechanisms in attention matrices
of one-layer Transformers. Our work stands as a significant stride in
unraveling their operation complexities, particularly in the realm of complex
algebraic tasks.
| [
{
"version": "v1",
"created": "Mon, 12 Feb 2024 05:52:06 GMT"
},
{
"version": "v2",
"created": "Fri, 24 May 2024 07:28:24 GMT"
},
{
"version": "v3",
"created": "Wed, 16 Oct 2024 06:48:42 GMT"
},
{
"version": "v4",
"created": "Sun, 9 Mar 2025 07:14:46 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Li",
"Chenyang",
""
],
[
"Liang",
"Yingyu",
""
],
[
"Shi",
"Zhenmei",
""
],
[
"Song",
"Zhao",
""
],
[
"Zhou",
"Tianyi",
""
]
]
| TITLE: Fourier Circuits in Neural Networks and Transformers: A Case Study of
Modular Arithmetic with Multiple Inputs
ABSTRACT: In the evolving landscape of machine learning, a pivotal challenge lies in
deciphering the internal representations harnessed by neural networks and
Transformers. Building on recent progress toward comprehending how networks
execute distinct target functions, our study embarks on an exploration of the
underlying reasons behind networks adopting specific computational strategies.
We direct our focus to the complex algebraic learning task of modular addition
involving $k$ inputs. Our research presents a thorough analytical
characterization of the features learned by stylized one-hidden layer neural
networks and one-layer Transformers in addressing this task. A cornerstone of
our theoretical framework is the elucidation of how the principle of margin
maximization shapes the features adopted by one-hidden layer neural networks.
Let $p$ denote the modulus, $D_p$ denote the dataset of modular arithmetic with
$k$ inputs and $m$ denote the network width. We demonstrate that a neuron count
of $ m \geq 2^{2k-2} \cdot (p-1) $, these networks attain a maximum $ L_{2,k+1}
$-margin on the dataset $ D_p $. Furthermore, we establish that each
hidden-layer neuron aligns with a specific Fourier spectrum, integral to
solving modular addition problems. By correlating our findings with the
empirical observations of similar studies, we contribute to a deeper
comprehension of the intrinsic computational mechanisms of neural networks.
Furthermore, we observe similar computational mechanisms in attention matrices
of one-layer Transformers. Our work stands as a significant stride in
unraveling their operation complexities, particularly in the realm of complex
algebraic tasks.
| no_new_dataset | 0.94366 |
2402.11345 | Nuojin Cheng | Nuojin Cheng and Stephen Becker | Variational Entropy Search for Adjusting Expected Improvement | This is a preliminary technical report. For a more comprehensive
study, please refer to arXiv:2501.18756 | null | null | null | stat.ML cs.LG math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Bayesian optimization is a widely used technique for optimizing black-box
functions, with Expected Improvement (EI) being the most commonly utilized
acquisition function in this domain. While EI is often viewed as distinct from
other information-theoretic acquisition functions, such as entropy search (ES)
and max-value entropy search (MES), our work reveals that EI can be considered
a special case of MES when approached through variational inference (VI). In
this context, we have developed the Variational Entropy Search (VES)
methodology and the VES-Gamma algorithm, which adapts EI by incorporating
principles from information-theoretic concepts. The efficacy of VES-Gamma is
demonstrated across a variety of test functions and read datasets, highlighting
its theoretical and practical utilities in Bayesian optimization scenarios.
| [
{
"version": "v1",
"created": "Sat, 17 Feb 2024 17:37:53 GMT"
},
{
"version": "v2",
"created": "Sun, 9 Mar 2025 15:29:40 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Cheng",
"Nuojin",
""
],
[
"Becker",
"Stephen",
""
]
]
| TITLE: Variational Entropy Search for Adjusting Expected Improvement
ABSTRACT: Bayesian optimization is a widely used technique for optimizing black-box
functions, with Expected Improvement (EI) being the most commonly utilized
acquisition function in this domain. While EI is often viewed as distinct from
other information-theoretic acquisition functions, such as entropy search (ES)
and max-value entropy search (MES), our work reveals that EI can be considered
a special case of MES when approached through variational inference (VI). In
this context, we have developed the Variational Entropy Search (VES)
methodology and the VES-Gamma algorithm, which adapts EI by incorporating
principles from information-theoretic concepts. The efficacy of VES-Gamma is
demonstrated across a variety of test functions and read datasets, highlighting
its theoretical and practical utilities in Bayesian optimization scenarios.
| no_new_dataset | 0.945349 |
2402.12767 | Zijian Li | Zijian Li, Ruichu Cai, Zhenhui Yang, Haiqin Huang, Guangyi Chen, Yifan
Shen, Zhengming Chen, Xiangchen Song, Kun Zhang | Nonstationary Time Series Forecasting via Unknown Distribution
Adaptation | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As environments evolve, temporal distribution shifts can degrade time series
forecasting performance. A straightforward solution is to adapt to
nonstationary changes while preserving stationary dependencies. Hence, some
methods disentangle stationary and nonstationary components by assuming uniform
distribution shifts, but it is impractical since when the distribution changes
is unknown. To address this challenge, we propose the \textbf{U}nknown
\textbf{D}istribution \textbf{A}daptation (\textbf{UDA}) model for
nonstationary time series forecasting, which detects when distribution shifts
occur and disentangles stationary/nonstationary latent variables, thus enabling
adaptation to unknown distribution without assuming a uniform distribution
shift. Specifically, under a Hidden Markov assumption of latent environments,
we demonstrate that the latent environments are identifiable. Sequentially, we
further disentangle stationary/nonstationary latent variables by leveraging the
variability of historical information. Based on these theoretical results, we
propose a variational autoencoder-based model, which incorporates an
autoregressive hidden Markov model to estimate latent environments.
Additionally, we further devise the modular prior networks to disentangle
stationary/nonstationary latent variables. These two modules realize automatic
adaptation and enhance nonstationary forecasting performance. Experimental
results on several datasets validate the effectiveness of our approach.
| [
{
"version": "v1",
"created": "Tue, 20 Feb 2024 07:16:12 GMT"
},
{
"version": "v2",
"created": "Sat, 13 Apr 2024 20:03:26 GMT"
},
{
"version": "v3",
"created": "Fri, 7 Jun 2024 11:11:31 GMT"
},
{
"version": "v4",
"created": "Mon, 10 Mar 2025 02:11:51 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Li",
"Zijian",
""
],
[
"Cai",
"Ruichu",
""
],
[
"Yang",
"Zhenhui",
""
],
[
"Huang",
"Haiqin",
""
],
[
"Chen",
"Guangyi",
""
],
[
"Shen",
"Yifan",
""
],
[
"Chen",
"Zhengming",
""
],
[
"Song",
"Xiangchen",
""
],
[
"Zhang",
"Kun",
""
]
]
| TITLE: Nonstationary Time Series Forecasting via Unknown Distribution
Adaptation
ABSTRACT: As environments evolve, temporal distribution shifts can degrade time series
forecasting performance. A straightforward solution is to adapt to
nonstationary changes while preserving stationary dependencies. Hence, some
methods disentangle stationary and nonstationary components by assuming uniform
distribution shifts, but it is impractical since when the distribution changes
is unknown. To address this challenge, we propose the \textbf{U}nknown
\textbf{D}istribution \textbf{A}daptation (\textbf{UDA}) model for
nonstationary time series forecasting, which detects when distribution shifts
occur and disentangles stationary/nonstationary latent variables, thus enabling
adaptation to unknown distribution without assuming a uniform distribution
shift. Specifically, under a Hidden Markov assumption of latent environments,
we demonstrate that the latent environments are identifiable. Sequentially, we
further disentangle stationary/nonstationary latent variables by leveraging the
variability of historical information. Based on these theoretical results, we
propose a variational autoencoder-based model, which incorporates an
autoregressive hidden Markov model to estimate latent environments.
Additionally, we further devise the modular prior networks to disentangle
stationary/nonstationary latent variables. These two modules realize automatic
adaptation and enhance nonstationary forecasting performance. Experimental
results on several datasets validate the effectiveness of our approach.
| no_new_dataset | 0.942507 |
2402.15183 | Zirui Guo | Zirui Guo, Lianghao Xia, Yanhua Yu, Yuling Wang, Kangkang Lu, Zhiyong
Huang, Chao Huang | GraphEdit: Large Language Models for Graph Structure Learning | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Graph Structure Learning (GSL) focuses on capturing intrinsic dependencies
and interactions among nodes in graph-structured data by generating novel graph
structures. Graph Neural Networks (GNNs) have emerged as promising GSL
solutions, utilizing recursive message passing to encode node-wise
inter-dependencies. However, many existing GSL methods heavily depend on
explicit graph structural information as supervision signals, leaving them
susceptible to challenges such as data noise and sparsity. In this work, we
propose GraphEdit, an approach that leverages large language models (LLMs) to
learn complex node relationships in graph-structured data. By enhancing the
reasoning capabilities of LLMs through instruction-tuning over graph
structures, we aim to overcome the limitations associated with explicit graph
structural information and enhance the reliability of graph structure learning.
Our approach not only effectively denoises noisy connections but also
identifies node-wise dependencies from a global perspective, providing a
comprehensive understanding of the graph structure. We conduct extensive
experiments on multiple benchmark datasets to demonstrate the effectiveness and
robustness of GraphEdit across various settings. We have made our model
implementation available at: https://github.com/HKUDS/GraphEdit.
| [
{
"version": "v1",
"created": "Fri, 23 Feb 2024 08:29:42 GMT"
},
{
"version": "v2",
"created": "Tue, 27 Feb 2024 08:22:11 GMT"
},
{
"version": "v3",
"created": "Thu, 29 Feb 2024 04:15:44 GMT"
},
{
"version": "v4",
"created": "Tue, 5 Mar 2024 05:22:00 GMT"
},
{
"version": "v5",
"created": "Mon, 10 Mar 2025 14:04:39 GMT"
}
]
| 2025-03-11T00:00:00 | [
[
"Guo",
"Zirui",
""
],
[
"Xia",
"Lianghao",
""
],
[
"Yu",
"Yanhua",
""
],
[
"Wang",
"Yuling",
""
],
[
"Lu",
"Kangkang",
""
],
[
"Huang",
"Zhiyong",
""
],
[
"Huang",
"Chao",
""
]
]
| TITLE: GraphEdit: Large Language Models for Graph Structure Learning
ABSTRACT: Graph Structure Learning (GSL) focuses on capturing intrinsic dependencies
and interactions among nodes in graph-structured data by generating novel graph
structures. Graph Neural Networks (GNNs) have emerged as promising GSL
solutions, utilizing recursive message passing to encode node-wise
inter-dependencies. However, many existing GSL methods heavily depend on
explicit graph structural information as supervision signals, leaving them
susceptible to challenges such as data noise and sparsity. In this work, we
propose GraphEdit, an approach that leverages large language models (LLMs) to
learn complex node relationships in graph-structured data. By enhancing the
reasoning capabilities of LLMs through instruction-tuning over graph
structures, we aim to overcome the limitations associated with explicit graph
structural information and enhance the reliability of graph structure learning.
Our approach not only effectively denoises noisy connections but also
identifies node-wise dependencies from a global perspective, providing a
comprehensive understanding of the graph structure. We conduct extensive
experiments on multiple benchmark datasets to demonstrate the effectiveness and
robustness of GraphEdit across various settings. We have made our model
implementation available at: https://github.com/HKUDS/GraphEdit.
| no_new_dataset | 0.951369 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.