id
string | submitter
string | authors
string | title
string | comments
string | journal-ref
string | doi
string | report-no
string | categories
string | license
string | abstract
string | versions
list | update_date
timestamp[s] | authors_parsed
sequence |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2506.02675 | Armin Ilg | Armin Ilg (1), Fabrizio Palla (2) (on behalf of FCC, (1) University of Zurich, (2) INFN Pisa) | Design, performance and future prospects of vertex detectors at the FCC-ee | 6 pages, 7 figures, presented at the 42nd International Conference on High Energy Physics (ICHEP2024) | PoS(ICHEP2024)1062 | 10.22323/1.476.1062 | null | hep-ex physics.ins-det | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The CERN proposed $e^+e^-$ Future Circular Collider (FCC-ee) is an electroweak, flavour, Higgs and top factory with unprecedented luminosities. Many measurements at the FCC-ee will rely on precisely determining the particle production vertices using dedicated vertex detectors. All vertex detector designs use Monolithic Active Pixel Sensors (MAPS) with a single-hit resolution of $\approx$ 3 $\mu$m and a material budget as low as 0.3% of a radiation length per detection layer, which is within specifications for most of the physics analyses. This contribution presents the status of the fully engineered vertex detectors and their integration with the collider beam pipe and discusses their predicted performance using the DD4hep full simulation. A concept for an ultra-light vertex detector using curved wafer-scale MAPS is also presented, which allows reducing the material budget by almost a factor of three. This improves the vertexing capabilities, especially for heavy flavour decays, such as $B^0 \rightarrow K^{*0} \tau^+ \tau^-$. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:20:12 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Ilg",
"Armin",
""
],
[
"Palla",
"Fabrizio",
""
]
] |
2506.02676 | Patrick Pfreundschuh | Patrick Pfreundschuh, Giovanni Cioffi, Cornelius von Einem, Alexander Wyss, Hans Wernher van de Venn, Cesar Cadena, Davide Scaramuzza, Roland Siegwart, and Alireza Darvishy | Sight Guide: A Wearable Assistive Perception and Navigation System for the Vision Assistance Race in the Cybathlon 2024 | null | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Visually impaired individuals face significant challenges navigating and interacting with unknown situations, particularly in tasks requiring spatial awareness and semantic scene understanding. To accelerate the development and evaluate the state of technologies that enable visually impaired people to solve these tasks, the Vision Assistance Race (VIS) at the Cybathlon 2024 competition was organized. In this work, we present Sight Guide, a wearable assistive system designed for the VIS. The system processes data from multiple RGB and depth cameras on an embedded computer that guides the user through complex, real-world-inspired tasks using vibration signals and audio commands. Our software architecture integrates classical robotics algorithms with learning-based approaches to enable capabilities such as obstacle avoidance, object detection, optical character recognition, and touchscreen interaction. In a testing environment, Sight Guide achieved a 95.7% task success rate, and further demonstrated its effectiveness during the Cybathlon competition. This work provides detailed insights into the system design, evaluation results, and lessons learned, and outlines directions towards a broader real-world applicability. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:22:02 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Pfreundschuh",
"Patrick",
""
],
[
"Cioffi",
"Giovanni",
""
],
[
"von Einem",
"Cornelius",
""
],
[
"Wyss",
"Alexander",
""
],
[
"van de Venn",
"Hans Wernher",
""
],
[
"Cadena",
"Cesar",
""
],
[
"Scaramuzza",
"Davide",
""
],
[
"Siegwart",
"Roland",
""
],
[
"Darvishy",
"Alireza",
""
]
] |
2506.02677 | Jintao Tong | Jintao Tong, Yixiong Zou, Guangyao Chen, Yuhua Li, Ruixuan Li | Self-Disentanglement and Re-Composition for Cross-Domain Few-Shot Segmentation | Accepted by ICML 2025 | null | null | null | cs.CV cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Cross-Domain Few-Shot Segmentation (CD-FSS) aims to transfer knowledge from a source-domain dataset to unseen target-domain datasets with limited annotations. Current methods typically compare the distance between training and testing samples for mask prediction. However, we find an entanglement problem exists in this widely adopted method, which tends to bind sourcedomain patterns together and make each of them hard to transfer. In this paper, we aim to address this problem for the CD-FSS task. We first find a natural decomposition of the ViT structure, based on which we delve into the entanglement problem for an interpretation. We find the decomposed ViT components are crossly compared between images in distance calculation, where the rational comparisons are entangled with those meaningless ones by their equal importance, leading to the entanglement problem. Based on this interpretation, we further propose to address the entanglement problem by learning to weigh for all comparisons of ViT components, which learn disentangled features and re-compose them for the CD-FSS task, benefiting both the generalization and finetuning. Experiments show that our model outperforms the state-of-the-art CD-FSS method by 1.92% and 1.88% in average accuracy under 1-shot and 5-shot settings, respectively. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:23:20 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Tong",
"Jintao",
""
],
[
"Zou",
"Yixiong",
""
],
[
"Chen",
"Guangyao",
""
],
[
"Li",
"Yuhua",
""
],
[
"Li",
"Ruixuan",
""
]
] |
2506.02678 | Zhong-Zhi Li | Zhong-Zhi Li, Xiao Liang, Zihao Tang, Lei Ji, Peijie Wang, Haotian Xu, Xing W, Haizhen Huang, Weiwei Deng, Ying Nian Wu, Yeyun Gong, Zhijiang Guo, Xiao Liu, Fei Yin, Cheng-Lin Liu | TL;DR: Too Long, Do Re-weighting for Effcient LLM Reasoning Compression | null | null | null | null | cs.CL cs.CE cs.NA math.NA | http://creativecommons.org/licenses/by/4.0/ | Large Language Models (LLMs) have recently achieved remarkable progress by leveraging Reinforcement Learning and extended Chain-of-Thought (CoT) techniques. However, the challenge of performing efficient language reasoning--especially during inference with extremely long outputs--has drawn increasing attention from the research community. In this work, we propose a dynamic ratio-based training pipeline that does not rely on sophisticated data annotations or interpolation between multiple models. We continuously balance the weights between the model's System-1 and System-2 data to eliminate redundant reasoning processes while preserving the model's reasoning capability. We validate our approach across models on DeepSeek-R1-Distill-7B and DeepSeek-R1-Distill-14B and on a diverse set of benchmarks with varying difficulty levels. Our method significantly reduces the number of output tokens by nearly 40% while maintaining the accuracy of the reasoning. Our code and data will be available soon. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:23:41 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Li",
"Zhong-Zhi",
""
],
[
"Liang",
"Xiao",
""
],
[
"Tang",
"Zihao",
""
],
[
"Ji",
"Lei",
""
],
[
"Wang",
"Peijie",
""
],
[
"Xu",
"Haotian",
""
],
[
"W",
"Xing",
""
],
[
"Huang",
"Haizhen",
""
],
[
"Deng",
"Weiwei",
""
],
[
"Wu",
"Ying Nian",
""
],
[
"Gong",
"Yeyun",
""
],
[
"Guo",
"Zhijiang",
""
],
[
"Liu",
"Xiao",
""
],
[
"Yin",
"Fei",
""
],
[
"Liu",
"Cheng-Lin",
""
]
] |
2506.02679 | Gabriele Digregorio | Gabriele Digregorio, Francesco Bleggi, Federico Caroli, Michele Carminati, Stefano Zanero, Stefano Longari | Poster: FedBlockParadox -- A Framework for Simulating and Securing Decentralized Federated Learning | International Conference on Detection of Intrusions and Malware, and Vulnerability Assessment (DIMVA '25), 2025 | null | null | null | cs.CR | http://creativecommons.org/licenses/by/4.0/ | A significant body of research in decentralized federated learning focuses on combining the privacy-preserving properties of federated learning with the resilience and transparency offered by blockchain-based systems. While these approaches are promising, they often lack flexible tools to evaluate system robustness under adversarial conditions. To fill this gap, we present FedBlockParadox, a modular framework for modeling and evaluating decentralized federated learning systems built on blockchain technologies, with a focus on resilience against a broad spectrum of adversarial attack scenarios. It supports multiple consensus protocols, validation methods, aggregation strategies, and configurable attack models. By enabling controlled experiments, FedBlockParadox provides a valuable resource for researchers developing secure, decentralized learning solutions. The framework is open-source and built to be extensible by the community. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:25:06 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Digregorio",
"Gabriele",
""
],
[
"Bleggi",
"Francesco",
""
],
[
"Caroli",
"Federico",
""
],
[
"Carminati",
"Michele",
""
],
[
"Zanero",
"Stefano",
""
],
[
"Longari",
"Stefano",
""
]
] |
2506.02680 | Julius Erbach | Julius Erbach, Dominik Narnhofer, Andreas Dombos, Bernt Schiele, Jan Eric Lenssen, Konrad Schindler | Solving Inverse Problems with FLAIR | null | null | null | null | cs.CV eess.IV | http://creativecommons.org/licenses/by/4.0/ | Flow-based latent generative models such as Stable Diffusion 3 are able to generate images with remarkable quality, even enabling photorealistic text-to-image generation. Their impressive performance suggests that these models should also constitute powerful priors for inverse imaging problems, but that approach has not yet led to comparable fidelity. There are several key obstacles: (i) the encoding into a lower-dimensional latent space makes the underlying (forward) mapping non-linear; (ii) the data likelihood term is usually intractable; and (iii) learned generative models struggle to recover rare, atypical data modes during inference. We present FLAIR, a novel training free variational framework that leverages flow-based generative models as a prior for inverse problems. To that end, we introduce a variational objective for flow matching that is agnostic to the type of degradation, and combine it with deterministic trajectory adjustments to recover atypical modes. To enforce exact consistency with the observed data, we decouple the optimization of the data fidelity and regularization terms. Moreover, we introduce a time-dependent calibration scheme in which the strength of the regularization is modulated according to off-line accuracy estimates. Results on standard imaging benchmarks demonstrate that FLAIR consistently outperforms existing diffusion- and flow-based methods in terms of reconstruction quality and sample diversity. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:29:47 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Erbach",
"Julius",
""
],
[
"Narnhofer",
"Dominik",
""
],
[
"Dombos",
"Andreas",
""
],
[
"Schiele",
"Bernt",
""
],
[
"Lenssen",
"Jan Eric",
""
],
[
"Schindler",
"Konrad",
""
]
] |
2506.02681 | Oubo You | Oubo You, Zhaoqi Jiang, Jinhui Shi, Qing Dai, Chunying Guan, Shuang Zhang | Adiabaticity violation under arbitrarily slow evolution | 14 pages, 3 figures | null | null | null | quant-ph physics.optics | http://creativecommons.org/licenses/by-nc-sa/4.0/ | The quantum adiabatic theorem, a cornerstone of quantum mechanics, asserts that a gapped quantum system remains in its instantaneous eigenstate during sufficiently slow evolution, provided no resonances occur. Here we challenge this principle and show that adiabaticity can be violated even in arbitrarily slow processes. We introduce two new parameters, Instantaneous Transition Accumulation (ITA) and Instantaneous Transition Probability (ITP), to redefine the framework of adiabatic evolution. These parameters, grounded in cross-Berry connections and eigenstate amplitudes, reveal the dynamic and geometric factors governing adiabaticity. Using a new Phase Difference Manipulation (PDM) method, we control ITP and ITA to induce adiabaticity violation in a Landau-Zener (LZ) process. We experimentally demonstrate this counterintuitive phenomenon in a photonic waveguide system, where a slow LZ process defies adiabaticity, switching energy levels despite a fivefold slower evolution speed than a conventional adiabatic process. This discovery reshapes our understanding of quantum evolution and holds potential for quantum computing, topological physics, and photonic technologies. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:29:58 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"You",
"Oubo",
""
],
[
"Jiang",
"Zhaoqi",
""
],
[
"Shi",
"Jinhui",
""
],
[
"Dai",
"Qing",
""
],
[
"Guan",
"Chunying",
""
],
[
"Zhang",
"Shuang",
""
]
] |
2506.02682 | In Dae Jong | Kazuhiro Ichihara, In Dae Jong | Cosmetic surgeries on knots in homology spheres and the Casson--Walker invariant | 10 pages | null | null | null | math.GT | http://creativecommons.org/licenses/by/4.0/ | We study cosmetic surgeries on a knot in a homology sphere. Several constraints on knots and surgery slopes to admit such surgeries are given. Our main ingredient is the rational surgery formula of the Casson--Walker invariant for 2-component links in the 3-sphere. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:32:41 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Ichihara",
"Kazuhiro",
""
],
[
"Jong",
"In Dae",
""
]
] |
2506.02683 | Zhengdong Lu | Zhengdong Lu, Weikai Lu, Yiling Tao, Yun Dai, ZiXuan Chen, Huiping Zhuang, Cen Chen, Hao Peng, Ziqian Zeng | Decompose, Plan in Parallel, and Merge: A Novel Paradigm for Large Language Models based Planning with Multiple Constraints | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Despite significant advances in Large Language Models (LLMs), planning tasks still present challenges for LLM-based agents. Existing planning methods face two key limitations: heavy constraints and cascading errors. To address these limitations, we propose a novel parallel planning paradigm, which Decomposes, Plans for subtasks in Parallel, and Merges subplans into a final plan (DPPM). Specifically, DPPM decomposes the complex task based on constraints into subtasks, generates the subplan for each subtask in parallel, and merges them into a global plan. In addition, our approach incorporates a verification and refinement module, enabling error correction and conflict resolution. Experimental results demonstrate that DPPM significantly outperforms existing methods in travel planning tasks. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:33:13 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Lu",
"Zhengdong",
""
],
[
"Lu",
"Weikai",
""
],
[
"Tao",
"Yiling",
""
],
[
"Dai",
"Yun",
""
],
[
"Chen",
"ZiXuan",
""
],
[
"Zhuang",
"Huiping",
""
],
[
"Chen",
"Cen",
""
],
[
"Peng",
"Hao",
""
],
[
"Zeng",
"Ziqian",
""
]
] |
2506.02684 | Meng Hock Koh | W. Asous, Mastura Syamimi Abdullah, Meng-Hock Koh, Kok-Siong Khoo | Deformed magic numbers at $N=$178 and $Z=$120, 124 in the 112 $\leq N \leq $ 190 superheavy region from Skyrme mean-field calculations | 15 figures, 10 tables | Published in the IJMPE 34:5, 2550020 (2025) | 10.1142/S021830132550020X | null | nucl-th | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Background: Various motivations for exploration of superheavy region revolve around the question on whether 126 is a spherical proton magic number, as is the case for neutrons. In exploring this region, identification of nuclei with relatively longer half-life as compared to its neighbours is crucial for experimental studies. Such information is provided from theoretical predictions, which are however, heavily dependent on the theoretical model used and observable quantities under investigation.
Purpose: Limiting ourselves to the Skyrme Hartree-Fock-plus-Bardeen-Cooper-Schrieffer approach, we aimed to analyse the appearance of a nuclear region with relatively high stability associated with emergence of spherical and deformed magic numbers in the region of $170 \le N \le 190$ ($112 \le Z \le 130$) based on various observables.
Methods: Three Skyrme parametrizations namely the SkM* frequently employed for fission calculations, and the SLy5 and SLy4 commonly used for superheavy region, are considered to provide comparisons within the Skyrme mean-field approach. We evaluated the variation of electric quadrupole deformation ($\beta_{20}$), binding energy per nucleon ($BE/A$), two-nucleon separation energy differential ($\delta S2_{q}$), alpha-decay energy ($Q_{\alpha}$) and alpha-decay half-lives ($T_{1/2}$).
Conclusion: Our analyses suggest that neutron number $N = 178$ is candidate for deformed magic number around proton number $114 \le Z \le 118$. For protons, $Z = 120$ and $124$ appears to be a candidate for deformed magic number at around $N = 172 \sim 178$. Both sets of deformed magic numbers appear at oblate ground-state deformation. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:33:25 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Asous",
"W.",
""
],
[
"Abdullah",
"Mastura Syamimi",
""
],
[
"Koh",
"Meng-Hock",
""
],
[
"Khoo",
"Kok-Siong",
""
]
] |
2506.02685 | Hohyun Kim | Hohyun Kim, Seunggeun Lee, Min-hwan Oh | Symmetry-Aware GFlowNets | 29 pages; Accepted at ICML 2025 | null | null | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Generative Flow Networks (GFlowNets) offer a powerful framework for sampling graphs in proportion to their rewards. However, existing approaches suffer from systematic biases due to inaccuracies in state transition probability computations. These biases, rooted in the inherent symmetries of graphs, impact both atom-based and fragment-based generation schemes. To address this challenge, we introduce Symmetry-Aware GFlowNets (SA-GFN), a method that incorporates symmetry corrections into the learning process through reward scaling. By integrating bias correction directly into the reward structure, SA-GFN eliminates the need for explicit state transition computations. Empirical results show that SA-GFN enables unbiased sampling while enhancing diversity and consistently generating high-reward graphs that closely match the target distribution. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:38:15 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Kim",
"Hohyun",
""
],
[
"Lee",
"Seunggeun",
""
],
[
"Oh",
"Min-hwan",
""
]
] |
2506.02686 | Stefano Guarino | Stefano Guarino, Davide Torre and Enrico Mastrostefano | Random Hyperbolic Graphs with Arbitrary Mesoscale Structures | null | null | null | null | cs.SI physics.app-ph physics.data-an | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Real-world networks exhibit universal structural properties such as sparsity, small-worldness, heterogeneous degree distributions, high clustering, and community structures. Geometric network models, particularly Random Hyperbolic Graphs (RHGs), effectively capture many of these features by embedding nodes in a latent similarity space. However, networks are often characterized by specific connectivity patterns between groups of nodes -- i.e. communities -- that are not geometric, in the sense that the dissimilarity between groups do not obey the triangle inequality. Structuring connections only based on the interplay of similarity and popularity thus poses fundamental limitations on the mesoscale structure of the networks that RHGs can generate. To address this limitation, we introduce the Random Hyperbolic Block Model (RHBM), which extends RHGs by incorporating block structures within a maximum-entropy framework. We demonstrate the advantages of the RHBM through synthetic network analyses, highlighting its ability to preserve community structures where purely geometric models fail. Our findings emphasize the importance of latent geometry in network modeling while addressing its limitations in controlling mesoscale mixing patterns. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:40:03 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Guarino",
"Stefano",
""
],
[
"Torre",
"Davide",
""
],
[
"Mastrostefano",
"Enrico",
""
]
] |
2506.02688 | Cornelius Buerkle | Peter Popov, Lorenzo Strigini, Cornelius Buerkle, Fabian Oboril, Michael Paulitsch | Stochastic Modeling of Road Hazards on Intersections and their Effect on Safety of Autonomous Vehicles | This work has been submitted to the IEEE for possible publication | null | null | null | cs.RO | http://creativecommons.org/licenses/by/4.0/ | Autonomous vehicles (AV) look set to become common on our roads within the next few years. However, to achieve the final breakthrough, not only functional progress is required, but also satisfactory safety assurance must be provided. Among those, a question demanding special attention is the need to assess and quantify the overall safety of an AV. Such an assessment must consider on the one hand the imperfections of the AV functionality and on the other hand its interaction with the environment. In a previous paper we presented a model-based approach to AV safety assessment in which we use a probabilistic model to describe road hazards together with the impact on AV safety of imperfect behavior of AV functions, such as safety monitors and perception systems. With this model, we are able to quantify the likelihood of the occurrence of a fatal accident, for a single operating condition. In this paper, we extend the approach and show how the model can deal explicitly with a set of different operating conditions defined in a given ODD. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:41:34 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Popov",
"Peter",
""
],
[
"Strigini",
"Lorenzo",
""
],
[
"Buerkle",
"Cornelius",
""
],
[
"Oboril",
"Fabian",
""
],
[
"Paulitsch",
"Michael",
""
]
] |
2506.02690 | Rebecca Zhao | Yurui Zhao, Xiang Wang, Jiahong Liu, Irwin King, Zhitao Huang | Towards Geometry Problem Solving in the Large Model Era: A Survey | 8pages, 4 figures, conference submission | null | null | null | cs.CV math.GT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Geometry problem solving (GPS) represents a critical frontier in artificial intelligence, with profound applications in education, computer-aided design, and computational graphics. Despite its significance, automating GPS remains challenging due to the dual demands of spatial understanding and rigorous logical reasoning. Recent advances in large models have enabled notable breakthroughs, particularly for SAT-level problems, yet the field remains fragmented across methodologies, benchmarks, and evaluation frameworks. This survey systematically synthesizes GPS advancements through three core dimensions: (1) benchmark construction, (2) textual and diagrammatic parsing, and (3) reasoning paradigms. We further propose a unified analytical paradigm, assess current limitations, and identify emerging opportunities to guide future research toward human-level geometric reasoning, including automated benchmark generation and interpretable neuro-symbolic integration. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:42:49 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Zhao",
"Yurui",
""
],
[
"Wang",
"Xiang",
""
],
[
"Liu",
"Jiahong",
""
],
[
"King",
"Irwin",
""
],
[
"Huang",
"Zhitao",
""
]
] |
2506.02691 | Yajun Ma | Li Liang, Yajun Ma and Gang Yang | Homological invariant properties under cleft extensions | 25pages | null | null | null | math.CT math.RA | http://creativecommons.org/licenses/by-sa/4.0/ | We study the behavior of the Gorenstein weak global dimension and Gorenstein flat-cotorsion global dimension under a cleft extension of rings; we prove that under some mild conditons the finiteness of the Gorenstein weak global dimension and that of the Gorenstein flat-cotorsion global dimension are both invariant. Moreover, we compare the relative singularity categories with respect to flat-cotorsion modules under a cleft extension of rings. Some applications to {\theta}-extensions and Morita context rings are given. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:42:51 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Liang",
"Li",
""
],
[
"Ma",
"Yajun",
""
],
[
"Yang",
"Gang",
""
]
] |
2506.02692 | Shu Yang | Shu Yang, Fengtao Zhou, Leon Mayer, Fuxiang Huang, Yiliang Chen, Yihui Wang, Sunan He, Yuxiang Nie, Xi Wang, \"Omer S\"umer, Yueming Jin, Huihui Sun, Shuchang Xu, Alex Qinyang Liu, Zheng Li, Jing Qin, Jeremy YuenChun Teoh, Lena Maier-Hein, Hao Chen | Large-scale Self-supervised Video Foundation Model for Intelligent Surgery | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Computer-Assisted Intervention (CAI) has the potential to revolutionize modern surgery, with surgical scene understanding serving as a critical component in supporting decision-making, improving procedural efficacy, and ensuring intraoperative safety. While existing AI-driven approaches alleviate annotation burdens via self-supervised spatial representation learning, their lack of explicit temporal modeling during pre-training fundamentally restricts the capture of dynamic surgical contexts, resulting in incomplete spatiotemporal understanding. In this work, we introduce the first video-level surgical pre-training framework that enables joint spatiotemporal representation learning from large-scale surgical video data. To achieve this, we constructed a large-scale surgical video dataset comprising 3,650 videos and approximately 3.55 million frames, spanning more than 20 surgical procedures and over 10 anatomical structures. Building upon this dataset, we propose SurgVISTA (Surgical Video-level Spatial-Temporal Architecture), a reconstruction-based pre-training method that captures intricate spatial structures and temporal dynamics through joint spatiotemporal modeling. Additionally, SurgVISTA incorporates image-level knowledge distillation guided by a surgery-specific expert to enhance the learning of fine-grained anatomical and semantic features. To validate its effectiveness, we established a comprehensive benchmark comprising 13 video-level datasets spanning six surgical procedures across four tasks. Extensive experiments demonstrate that SurgVISTA consistently outperforms both natural- and surgical-domain pre-trained models, demonstrating strong potential to advance intelligent surgical systems in clinically meaningful scenarios. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:42:54 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Yang",
"Shu",
""
],
[
"Zhou",
"Fengtao",
""
],
[
"Mayer",
"Leon",
""
],
[
"Huang",
"Fuxiang",
""
],
[
"Chen",
"Yiliang",
""
],
[
"Wang",
"Yihui",
""
],
[
"He",
"Sunan",
""
],
[
"Nie",
"Yuxiang",
""
],
[
"Wang",
"Xi",
""
],
[
"Sümer",
"Ömer",
""
],
[
"Jin",
"Yueming",
""
],
[
"Sun",
"Huihui",
""
],
[
"Xu",
"Shuchang",
""
],
[
"Liu",
"Alex Qinyang",
""
],
[
"Li",
"Zheng",
""
],
[
"Qin",
"Jing",
""
],
[
"Teoh",
"Jeremy YuenChun",
""
],
[
"Maier-Hein",
"Lena",
""
],
[
"Chen",
"Hao",
""
]
] |
2506.02693 | Sabir M. Gusein-Zade | Antonio Campillo, Felix Delgado, Sabir Gusein-Zade | Poincar\'e series of valuations on functions over a field | null | null | null | null | math.AG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | For a subfield $\K$ of the field $\C$ of complex numbers, we consider curve and divisorial valuations on the algebra $\K[[x,y]]$ of formal power series in two variables with the coeficients in $\K$. We compute the semigroup Poincar\'e series and the classical Poincar\'e series of a valuation. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:43:34 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Campillo",
"Antonio",
""
],
[
"Delgado",
"Felix",
""
],
[
"Gusein-Zade",
"Sabir",
""
]
] |
2506.02694 | Daichi Kimura | Daichi Kimura, Tomonori Izumitani, Hisashi Kashima | XicorAttention: Time Series Transformer Using Attention with Nonlinear Correlation | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Various Transformer-based models have been proposed for time series forecasting. These models leverage the self-attention mechanism to capture long-term temporal or variate dependencies in sequences. Existing methods can be divided into two approaches: (1) reducing computational cost of attention by making the calculations sparse, and (2) reshaping the input data to aggregate temporal features. However, existing attention mechanisms may not adequately capture inherent nonlinear dependencies present in time series data, leaving room for improvement. In this study, we propose a novel attention mechanism based on Chatterjee's rank correlation coefficient, which measures nonlinear dependencies between variables. Specifically, we replace the matrix multiplication in standard attention mechanisms with this rank coefficient to measure the query-key relationship. Since computing Chatterjee's correlation coefficient involves sorting and ranking operations, we introduce a differentiable approximation employing SoftSort and SoftRank. Our proposed mechanism, ``XicorAttention,'' integrates it into several state-of-the-art Transformer models. Experimental results on real-world datasets demonstrate that incorporating nonlinear correlation into the attention improves forecasting accuracy by up to approximately 9.1\% compared to existing models. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:43:45 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Kimura",
"Daichi",
""
],
[
"Izumitani",
"Tomonori",
""
],
[
"Kashima",
"Hisashi",
""
]
] |
2506.02696 | Jinyuan Luo | Jinyuan Luo, Zhen Fang, Yixuan Li, Seongheon Park, Ling Chen | Shaking to Reveal: Perturbation-Based Detection of LLM Hallucinations | null | null | null | null | cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Hallucination remains a key obstacle to the reliable deployment of large language models (LLMs) in real-world question answering tasks. A widely adopted strategy to detect hallucination, known as self-assessment, relies on the model's own output confidence to estimate the factual accuracy of its answers. However, this strategy assumes that the model's output distribution closely reflects the true data distribution, which may not always hold in practice. As bias accumulates through the model's layers, the final output can diverge from the underlying reasoning process, making output-level confidence an unreliable signal for hallucination detection. In this work, we propose Sample-Specific Prompting (SSP), a new framework that improves self-assessment by analyzing perturbation sensitivity at intermediate representations. These representations, being less influenced by model bias, offer a more faithful view of the model's latent reasoning process. Specifically, SSP dynamically generates noise prompts for each input and employs a lightweight encoder to amplify the changes in representations caused by the perturbation. A contrastive distance metric is then used to quantify these differences and separate truthful from hallucinated responses. By leveraging the dynamic behavior of intermediate representations under perturbation, SSP enables more reliable self-assessment. Extensive experiments demonstrate that SSP significantly outperforms prior methods across a range of hallucination detection benchmarks. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:44:28 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Luo",
"Jinyuan",
""
],
[
"Fang",
"Zhen",
""
],
[
"Li",
"Yixuan",
""
],
[
"Park",
"Seongheon",
""
],
[
"Chen",
"Ling",
""
]
] |
2506.02697 | Yuxuan Wu | Yuxuan Wu, Le Wang, Sanping Zhou, Mengnan Liu, Gang Hua, Haoxiang Li | LayoutRAG: Retrieval-Augmented Model for Content-agnostic Conditional Layout Generation | 12 pages, 5 figures | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Controllable layout generation aims to create plausible visual arrangements of element bounding boxes within a graphic design according to certain optional constraints, such as the type or position of a specific component. While recent diffusion or flow-matching models have achieved considerable advances in multifarious conditional generation tasks, there remains considerable room for generating optimal arrangements under given conditions. In this work, we propose to carry out layout generation through retrieving by conditions and reference-guided generation. Specifically, we retrieve appropriate layout templates according to given conditions as references. The references are then utilized to guide the denoising or flow-based transport process. By retrieving layouts compatible with the given conditions, we can uncover the potential information not explicitly provided in the given condition. Such an approach offers more effective guidance to the model during the generation process, in contrast to previous models that feed the condition to the model and let the model infer the unprovided layout attributes directly. Meanwhile, we design a condition-modulated attention that selectively absorbs retrieval knowledge, adapting to the difference between retrieved templates and given conditions. Extensive experiment results show that our method successfully produces high-quality layouts that meet the given conditions and outperforms existing state-of-the-art models. Code will be released upon acceptance. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:47:03 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Wu",
"Yuxuan",
""
],
[
"Wang",
"Le",
""
],
[
"Zhou",
"Sanping",
""
],
[
"Liu",
"Mengnan",
""
],
[
"Hua",
"Gang",
""
],
[
"Li",
"Haoxiang",
""
]
] |
2506.02698 | Yunhong Lu | Yunhong Lu, Qichao Wang, Hengyuan Cao, Xiaoyin Xu, Min Zhang | Smoothed Preference Optimization via ReNoise Inversion for Aligning Diffusion Models with Varied Human Preferences | Accepted by ICML 2025 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Direct Preference Optimization (DPO) aligns text-to-image (T2I) generation models with human preferences using pairwise preference data. Although substantial resources are expended in collecting and labeling datasets, a critical aspect is often neglected: \textit{preferences vary across individuals and should be represented with more granularity.} To address this, we propose SmPO-Diffusion, a novel method for modeling preference distributions to improve the DPO objective, along with a numerical upper bound estimation for the diffusion optimization objective. First, we introduce a smoothed preference distribution to replace the original binary distribution. We employ a reward model to simulate human preferences and apply preference likelihood averaging to improve the DPO loss, such that the loss function approaches zero when preferences are similar. Furthermore, we utilize an inversion technique to simulate the trajectory preference distribution of the diffusion model, enabling more accurate alignment with the optimization objective. Our approach effectively mitigates issues of excessive optimization and objective misalignment present in existing methods through straightforward modifications. Our SmPO-Diffusion achieves state-of-the-art performance in preference evaluation, outperforming baselines across metrics with lower training costs. The project page is https://jaydenlyh.github.io/SmPO-project-page/. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:47:22 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Lu",
"Yunhong",
""
],
[
"Wang",
"Qichao",
""
],
[
"Cao",
"Hengyuan",
""
],
[
"Xu",
"Xiaoyin",
""
],
[
"Zhang",
"Min",
""
]
] |
2506.02699 | Aleksandar Miladinovi\'c | Aleksandar Miladinovi\'c | Universal property of framed $G$-disc algebras | null | null | null | null | math.AT math.CT | http://creativecommons.org/licenses/by/4.0/ | Given a compact Lie group $G$ and its finite subgroup $H$ we prove that the $\infty$-category of $G/H$-framed $G$-disc algebras taking values in a $G$-symmetric monoidal category $\underline{\mathcal{C}}^{\otimes}$ is equivalent to the $\infty$-category of $V$-framed $H$-disc algebras (where $V$ is an $H$-representation) which take values in $\underline{\mathcal{C}}^{\otimes}_H$, the underlying $H$-symmetric monoidal subcategory of $\underline{\mathcal{C}}^{\otimes}$. We will use this construction to refine the $C_2$-action on the real topological Hochschild homology to an $O(2)$-action. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:49:24 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Miladinović",
"Aleksandar",
""
]
] |
2506.02700 | Zhengyang Li | Zhengyang Li and Hailin Deng | Cognitive Load-Driven VR Memory Palaces: Personalizing Focus and Recall Enhancement | 10 pages, 5 figures, submitted to HCII 2025. Includes EEG-based VR adaptation experiments, dynamic spatial modeling, and empirical evaluation of memory performance | null | null | null | cs.HC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Cognitive load, which varies across individuals, can significantly affect focus and memory performance.This study explores the integration of Virtual Reality (VR) with memory palace techniques, aiming to optimize VR environments tailored to individual cognitive load levels to improve focus and memory. We utilized EEG devices, specifically the Oculus Quest 2, to monitor Beta wave activity in 10 participants.By modeling their cognitive load profiles through polynomial regression, we dynamically adjusted spatial variables within a VR environment using Grasshopper, creating personalized experiences. Results indicate that 8 participants showed a notable increase in Beta wave activity, demonstrating improved focus and cognitive performance in the customized VR settings.These findings underscore the potential of VR-based memory environments, driven by cognitive load considerations, and provide valuable insights for advancing VR memory research | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:52:56 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Li",
"Zhengyang",
""
],
[
"Deng",
"Hailin",
""
]
] |
2506.02702 | Tibor Kubik | Tibor Kub\'ik, Fran\c{c}ois Guibault, Michal \v{S}pan\v{e}l, Herv\'e Lombaert | ToothForge: Automatic Dental Shape Generation using Synchronized Spectral Embeddings | Information Processing in Medical Imaging (IPMI2025) | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | We introduce ToothForge, a spectral approach for automatically generating novel 3D teeth, effectively addressing the sparsity of dental shape datasets. By operating in the spectral domain, our method enables compact machine learning modeling, allowing the generation of high-resolution tooth meshes in milliseconds. However, generating shape spectra comes with the instability of the decomposed harmonics. To address this, we propose modeling the latent manifold on synchronized frequential embeddings. Spectra of all data samples are aligned to a common basis prior to the training procedure, effectively eliminating biases introduced by the decomposition instability. Furthermore, synchronized modeling removes the limiting factor imposed by previous methods, which require all shapes to share a common fixed connectivity. Using a private dataset of real dental crowns, we observe a greater reconstruction quality of the synthetized shapes, exceeding those of models trained on unaligned embeddings. We also explore additional applications of spectral analysis in digital dentistry, such as shape compression and interpolation. ToothForge facilitates a range of approaches at the intersection of spectral analysis and machine learning, with fewer restrictions on mesh structure. This makes it applicable for shape analysis not only in dentistry, but also in broader medical applications, where guaranteeing consistent connectivity across shapes from various clinics is unrealistic. The code is available at https://github.com/tiborkubik/toothForge. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:56:22 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Kubík",
"Tibor",
""
],
[
"Guibault",
"François",
""
],
[
"Španěl",
"Michal",
""
],
[
"Lombaert",
"Hervé",
""
]
] |
2506.02703 | Khizar Hayat | Khizar Hayat and Baptiste Magnier | Data Leakage and Deceptive Performance: A Critical Examination of Credit Card Fraud Detection Methodologies | null | null | null | null | cs.LG cs.AI cs.CY | http://creativecommons.org/publicdomain/zero/1.0/ | This study critically examines the methodological rigor in credit card fraud detection research, revealing how fundamental evaluation flaws can overshadow algorithmic sophistication. Through deliberate experimentation with improper evaluation protocols, we demonstrate that even simple models can achieve deceptively impressive results when basic methodological principles are violated. Our analysis identifies four critical issues plaguing current approaches: (1) pervasive data leakage from improper preprocessing sequences, (2) intentional vagueness in methodological reporting, (3) inadequate temporal validation for transaction data, and (4) metric manipulation through recall optimization at precision's expense. We present a case study showing how a minimal neural network architecture with data leakage outperforms many sophisticated methods reported in literature, achieving 99.9\% recall despite fundamental evaluation flaws. These findings underscore that proper evaluation methodology matters more than model complexity in fraud detection research. The study serves as a cautionary example of how methodological rigor must precede architectural sophistication, with implications for improving research practices across machine learning applications. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:56:43 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Hayat",
"Khizar",
""
],
[
"Magnier",
"Baptiste",
""
]
] |
2506.02704 | Bastien Auvray | Bastien Auvray, Julien David, Richard Groult, Thierry Lecroq | Cartesian Forest Matching | Submitted to SPIRE 2025 | null | null | null | cs.DS | http://creativecommons.org/licenses/by/4.0/ | In this paper, we introduce the notion of Cartesian Forest, which generalizes Cartesian Trees, in order to deal with partially ordered sequences. We show that algorithms that solve both exact and approximate Cartesian Tree Matching can be adapted to solve Cartesian Forest Matching in average linear time. We adapt the notion of Cartesian Tree Signature to Cartesian Forests and show how filters can be used to experimentally improve the algorithm for the exact matching. We also show a one to one correspondence between Cartesian Forests and Schr\"oder Trees. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:57:28 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Auvray",
"Bastien",
""
],
[
"David",
"Julien",
""
],
[
"Groult",
"Richard",
""
],
[
"Lecroq",
"Thierry",
""
]
] |
2506.02705 | Po-Chieh Yu | Po-Chieh Yu | Residual 1D CNN for Low SFR Surface Density Regression: A Design Note | null | null | null | null | astro-ph.IM | http://creativecommons.org/licenses/by-nc-sa/4.0/ | This technical note describes the design and modular implementation of a one-dimensional convolutional neural network (1D CNN) adapted from residual networks (ResNet), developed for photometric regression tasks with an emphasis on low star formation rate surface density ($\Sigma_{\mathrm{SFR}}$) inference. The model features residual block structures optimized for sparse targets, with optional loss weighting and diagnostic tools for analyzing residual behavior. The implementation (version \texttt{v1.4}) originated during a collaborative project and is documented here independently. No external data are reproduced or analyzed. This note provides a reusable architectural reference for scalar regression problems in astronomy and related domains. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:57:28 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Yu",
"Po-Chieh",
""
]
] |
2506.02706 | Sajan Maharjan | Angelo Josey Caldeira, Sajan Maharjan, Srijoni Majumdar and Evangelos Pournaras | Collective Intelligence Outperforms Individual Talent: A Case Study in League of Legends | null | null | null | null | cs.SI | http://creativecommons.org/licenses/by/4.0/ | Gaming environments are popular testbeds for studying human interactions and behaviors in complex artificial intelligence systems. Particularly, in multiplayer online battle arena (MOBA) games, individuals collaborate in virtual environments of high realism that involves real-time strategic decision-making and trade-offs on resource management, information collection and sharing, team synergy and collective dynamics. This paper explores whether collective intelligence, emerging from cooperative behaviours exhibited by a group of individuals, who are not necessarily skillful but effectively engage in collaborative problem-solving tasks, exceeds individual intelligence observed within skillful individuals. This is shown via a case study in League of Legends, using machine learning algorithms and statistical methods applied to large-scale data collected for the same purpose. By modelling systematically game-specific metrics but also new game-agnostic topological and graph spectra measures of cooperative interactions, we demonstrate compelling insights about the superior performance of collective intelligence. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:03:43 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Caldeira",
"Angelo Josey",
""
],
[
"Maharjan",
"Sajan",
""
],
[
"Majumdar",
"Srijoni",
""
],
[
"Pournaras",
"Evangelos",
""
]
] |
2506.02707 | Junyi Tao | Junyi Tao, Ran Li, Salvador Pineda | Unit Commitment with Cost-Oriented Temporal Resolution | null | null | null | null | eess.SY cs.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Time-adaptive unit commitment (UC) has recently been investigated to reduce the scheduling costs by flexibly varying the temporal resolution, which is usually determined by clustering the net load patterns. However, there exists a misalignment between cost and net load patterns due to the discrete start-up costs and out-of-merit-order dispatch triggered by ramping and other constraints. The optimal time-adaptive resolution cannot be completely captured by clustering-based method. This paper proposes a cost-oriented method to address this misalignment by a novel bilevel optimization approach that is efficiently solved through a heuristic greedy algorithm. The impact of varying temporal resolution on the final scheduling costs are tested, based on which the temporal resolution is heuristically updated, achieving significant cost reduction without increasing the number of temporal periods. Subsequently, an improved discretized Adam optimization method together with offline warm start and online refinement strategy is proposed to efficiently search for the better temporal resolution configuration. Results show that the proposed cost-oriented UC temporal resolution determination method achieves enhanced cost efficiency. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:04:08 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Tao",
"Junyi",
""
],
[
"Li",
"Ran",
""
],
[
"Pineda",
"Salvador",
""
]
] |
2506.02708 | Naoto Tanji | Naoto Tanji and Toshihiko Yamasaki | Iterative Self-Improvement of Vision Language Models for Image Scoring and Self-Explanation | Accepted to ICIP2025 | null | null | null | cs.CV cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Image scoring is a crucial task in numerous real-world applications. To trust a model's judgment, understanding its rationale is essential. This paper proposes a novel training method for Vision Language Models (VLMs) to generate not only image scores but also corresponding justifications in natural language. Leveraging only an image scoring dataset and an instruction-tuned VLM, our method enables self-training, utilizing the VLM's generated text without relying on external data or models. In addition, we introduce a simple method for creating a dataset designed to improve alignment between predicted scores and their textual justifications. By iteratively training the model with Direct Preference Optimization on two distinct datasets and merging them, we can improve both scoring accuracy and the coherence of generated explanations. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:04:19 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Tanji",
"Naoto",
""
],
[
"Yamasaki",
"Toshihiko",
""
]
] |
2506.02709 | Vanessa Sochat | Vanessa Sochat, Daniel Milroy, Abhik Sarkar, Aniruddha Marathe | Usability Evaluation of Cloud for HPC Applications | 13 pages, 8 figures, 4 tables | null | null | null | cs.DC cs.PF | http://creativecommons.org/licenses/by/4.0/ | The rise of AI and the economic dominance of cloud computing have created a new nexus of innovation for high performance computing (HPC), which has a long history of driving scientific discovery. In addition to performance needs, scientific workflows increasingly demand capabilities of cloud environments: portability, reproducibility, dynamism, and automation. As converged cloud environments emerge, there is growing need to study their fit for HPC use cases. Here we present a cross-platform usability study that assesses 11 different HPC proxy applications and benchmarks across three clouds (Microsoft Azure, Amazon Web Services, and Google Cloud), six environments, and two compute configurations (CPU and GPU) against on-premises HPC clusters at a major center. We perform scaling tests of applications in all environments up to 28,672 CPUs and 256 GPUs. We present methodology and results to guide future study and provide a foundation to define best practices for running HPC workloads in cloud. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:05:30 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Sochat",
"Vanessa",
""
],
[
"Milroy",
"Daniel",
""
],
[
"Sarkar",
"Abhik",
""
],
[
"Marathe",
"Aniruddha",
""
]
] |
2506.02710 | Tim Nisslbeck | T. N. Nisslbeck, Wouter M. Kouw | Online Bayesian system identification in multivariate autoregressive models via message passing | 6 pages, 1 figure, conference: ECC2025 | null | null | null | eess.SP cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose a recursive Bayesian estimation procedure for multivariate autoregressive models with exogenous inputs based on message passing in a factor graph. Unlike recursive least-squares, our method produces full posterior distributions for both the autoregressive coefficients and noise precision. The uncertainties regarding these estimates propagate into the uncertainties on predictions for future system outputs, and support online model evidence calculations. We demonstrate convergence empirically on a synthetic autoregressive system and competitive performance on a double mass-spring-damper system. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:06:29 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Nisslbeck",
"T. N.",
""
],
[
"Kouw",
"Wouter M.",
""
]
] |
2506.02711 | Jing Xue | Jing Xue, Zhishen Sun, Haishan Ye, Luo Luo, Xiangyu Chang, Ivor Tsang, Guang Dai | Privacy Leaks by Adversaries: Adversarial Iterations for Membership Inference Attack | null | null | null | null | cs.CR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Membership inference attack (MIA) has become one of the most widely used and effective methods for evaluating the privacy risks of machine learning models. These attacks aim to determine whether a specific sample is part of the model's training set by analyzing the model's output. While traditional membership inference attacks focus on leveraging the model's posterior output, such as confidence on the target sample, we propose IMIA, a novel attack strategy that utilizes the process of generating adversarial samples to infer membership. We propose to infer the member properties of the target sample using the number of iterations required to generate its adversarial sample. We conduct experiments across multiple models and datasets, and our results demonstrate that the number of iterations for generating an adversarial sample is a reliable feature for membership inference, achieving strong performance both in black-box and white-box attack scenarios. This work provides a new perspective for evaluating model privacy and highlights the potential of adversarial example-based features for privacy leakage assessment. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:09:24 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Xue",
"Jing",
""
],
[
"Sun",
"Zhishen",
""
],
[
"Ye",
"Haishan",
""
],
[
"Luo",
"Luo",
""
],
[
"Chang",
"Xiangyu",
""
],
[
"Tsang",
"Ivor",
""
],
[
"Dai",
"Guang",
""
]
] |
2506.02712 | Jayadev Naram | Jayadev Naram, Fredrik Hellstr\"om, Ziming Wang, Rebecka J\"ornsten, Giuseppe Durisi | Theoretical Performance Guarantees for Partial Domain Adaptation via Partial Optimal Transport | ICML 2025 | null | null | null | cs.LG stat.ML | http://creativecommons.org/licenses/by-nc-sa/4.0/ | In many scenarios of practical interest, labeled data from a target distribution are scarce while labeled data from a related source distribution are abundant. One particular setting of interest arises when the target label space is a subset of the source label space, leading to the framework of partial domain adaptation (PDA). Typical approaches to PDA involve minimizing a domain alignment term and a weighted empirical loss on the source data, with the aim of transferring knowledge between domains. However, a theoretical basis for this procedure is lacking, and in particular, most existing weighting schemes are heuristic. In this work, we derive generalization bounds for the PDA problem based on partial optimal transport. These bounds corroborate the use of the partial Wasserstein distance as a domain alignment term, and lead to theoretically motivated explicit expressions for the empirical source loss weights. Inspired by these bounds, we devise a practical algorithm for PDA, termed WARMPOT. Through extensive numerical experiments, we show that WARMPOT is competitive with recent approaches, and that our proposed weights improve on existing schemes. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:09:28 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Naram",
"Jayadev",
""
],
[
"Hellström",
"Fredrik",
""
],
[
"Wang",
"Ziming",
""
],
[
"Jörnsten",
"Rebecka",
""
],
[
"Durisi",
"Giuseppe",
""
]
] |
2506.02713 | Xiaochong Lan | Xiaochong Lan, Jie Feng, Yizhou Sun, Chen Gao, Jiahuan Lei, Xinlei Shi, Hengliang Luo, Yong Li | Open-Set Living Need Prediction with Large Language Models | ACL 2025 Findings | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | Living needs are the needs people generate in their daily lives for survival and well-being. On life service platforms like Meituan, user purchases are driven by living needs, making accurate living need predictions crucial for personalized service recommendations. Traditional approaches treat this prediction as a closed-set classification problem, severely limiting their ability to capture the diversity and complexity of living needs. In this work, we redefine living need prediction as an open-set classification problem and propose PIGEON, a novel system leveraging large language models (LLMs) for unrestricted need prediction. PIGEON first employs a behavior-aware record retriever to help LLMs understand user preferences, then incorporates Maslow's hierarchy of needs to align predictions with human living needs. For evaluation and application, we design a recall module based on a fine-tuned text embedding model that links flexible need descriptions to appropriate life services. Extensive experiments on real-world datasets demonstrate that PIGEON significantly outperforms closed-set approaches on need-based life service recall by an average of 19.37%. Human evaluation validates the reasonableness and specificity of our predictions. Additionally, we employ instruction tuning to enable smaller LLMs to achieve competitive performance, supporting practical deployment. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:10:19 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Lan",
"Xiaochong",
""
],
[
"Feng",
"Jie",
""
],
[
"Sun",
"Yizhou",
""
],
[
"Gao",
"Chen",
""
],
[
"Lei",
"Jiahuan",
""
],
[
"Shi",
"Xinlei",
""
],
[
"Luo",
"Hengliang",
""
],
[
"Li",
"Yong",
""
]
] |
2506.02714 | Valeria Zitz | Valeria Zitz, Michael K\"uttner, Jonas Hummel, Michael T. Knierim, Michael Beigl, Tobias R\"oddiger | Heatables: Effects of Infrared-LED-Induced Ear Heating on Thermal Perception, Comfort, and Cognitive Performance | null | null | null | null | cs.HC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Maintaining thermal comfort in shared indoor environments remains challenging, as centralized HVAC systems are slow to adapt and standardized to group norms. Cold exposure not only reduces subjective comfort but can impair cognitive performance, particularly under moderate to severe cold stress. Personal Comfort Systems (PCS) have shown promise by providing localized heating, yet many designs target distal body parts with low thermosensitivity and often lack portability. In this work, we investigate whether targeted thermal stimulation using in-ear worn devices can manipulate thermal perception and enhance thermal comfort. We present Heatables, a novel in-ear wearable that emits Near-Infrared (NIR) and Infrared (IR) radiation via integrated LEDs to deliver localized optical heating. This approach leverages NIR-IR's ability to penetrate deeper tissues, offering advantages over traditional resistive heating limited to surface warming. In a placebo-controlled study with 24 participants, each exposed for 150 minutes in a cool office environment (approximately 17.5 degrees Celsius) to simulate sustained cold stress during typical sedentary office activities, Heatables significantly increased the perceived ambient temperature by around 1.5 degrees Celsius and delayed cold discomfort. Importantly, thermal benefits extended beyond the ear region, improving both whole-body comfort and thermal acceptability. These findings position in-ear NIR-IR-LED-based stimulation as a promising modality for unobtrusive thermal comfort enhancement in everyday contexts. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:12:16 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Zitz",
"Valeria",
""
],
[
"Küttner",
"Michael",
""
],
[
"Hummel",
"Jonas",
""
],
[
"Knierim",
"Michael T.",
""
],
[
"Beigl",
"Michael",
""
],
[
"Röddiger",
"Tobias",
""
]
] |
2506.02715 | Michael K\"uttner | Michael K\"uttner, Valeria Sitz, Kathrin Gerling, Michael Beigl, Tobias R\"oddiger | UltrasonicSpheres: Localized, Multi-Channel Sound Spheres Using Off-the-Shelf Speakers and Earables | null | null | null | null | cs.SD cs.HC eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a demo ofUltrasonicSpheres, a novel system for location-specific audio delivery using wearable earphones that decode ultrasonic signals into audible sound. Unlike conventional beamforming setups, UltrasonicSpheres relies on single ultrasonic speakers to broadcast localized audio with multiple channels, each encoded on a distinct ultrasonic carrier frequency. Users wearing our acoustically transparent earphones can demodulate their selected stream, such as exhibit narrations in a chosen language, while remaining fully aware of ambient environmental sounds. The experience preserves spatial audio perception, giving the impression that the sound originates directly from the physical location of the source. This enables personalized, localized audio without requiring pairing, tracking, or additional infrastructure. Importantly, visitors not equipped with the earphones are unaffected, as the ultrasonic signals are inaudible to the human ear. Our demo invites participants to explore multiple co-located audio zones and experience how UltrasonicSpheres supports unobtrusive delivery of personalized sound in public spaces. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:13:24 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Küttner",
"Michael",
""
],
[
"Sitz",
"Valeria",
""
],
[
"Gerling",
"Kathrin",
""
],
[
"Beigl",
"Michael",
""
],
[
"Röddiger",
"Tobias",
""
]
] |
2506.02716 | Sudin Ganguly | Sourav Karmakar, Sudin Ganguly, and Santanu K. Maiti | Reentrant localization in a quasiperiodic chain with correlated hopping sequences | 7 pages, 5 figures. Comments are welcome | null | null | null | cond-mat.dis-nn cond-mat.mes-hall cond-mat.str-el physics.comp-ph physics.data-an | http://creativecommons.org/licenses/by/4.0/ | Quasiperiodic systems are known to exhibit localization transitions in low dimensions, wherein all electronic states become localized beyond a critical disorder strength. Interestingly, recent studies have uncovered a reentrant localization (RL) phenomenon: upon further increasing the quasiperiodic disorder strength beyond the localization threshold, a subset of previously localized states can become delocalized again within a specific parameter window. While RL transitions have been primarily explored in systems with simple periodic modulations, such as dimerized or long-range hopping integrals, the impact of more intricate or correlated hopping structures on RL behavior remains largely elusive. In this work, we investigate the localization behavior in a one-dimensional lattice featuring staggered, correlated on-site potentials following the Aubry-Andr\'{e}-Harper model, along with off-diagonal hopping modulations structured according to quasiperiodic Fibonacci and Bronze Mean sequences. By systematically analyzing the fractal dimension, inverse participation ratio, and normalized participation ratio, we demonstrate the occurrence of RL transitions induced purely by the interplay between quasiperiodic on-site disorder and correlated hopping. Our findings highlight the crucial role of underlying structural correlations in governing localization-delocalization transitions in low-dimensional quasiperiodic systems, where the correlated disorder manifests in both diagonal and off-diagonal terms. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:15:19 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Karmakar",
"Sourav",
""
],
[
"Ganguly",
"Sudin",
""
],
[
"Maiti",
"Santanu K.",
""
]
] |
2506.02717 | Sergey Vyatchanin | A.A. Movsisian and A.I. Nazmiev and A.B. Matsko and S.P. Vyatchanin | Noiseless signal amplification in an opto-mechanical transducer | 5 pages, 4 figures | null | null | null | quant-ph | http://creativecommons.org/publicdomain/zero/1.0/ | The high-sensitivity quantum detection of a resonant classical force acting on a quantum oscillator can be substantially enhanced through the use of a resonant optical parametric transducer. We demonstrate that this approach not only enables quantum back-action evasion measurements that exceed the Standard Quantum Limit of sensitivity but also facilitates the noiseless amplification of the classical signal. This amplification is achieved by independently measuring the two modulation sidebands generated by the signal force, allowing for a more precise and noise-resistant detection process. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:15:23 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Movsisian",
"A. A.",
""
],
[
"Nazmiev",
"A. I.",
""
],
[
"Matsko",
"A. B.",
""
],
[
"Vyatchanin",
"S. P.",
""
]
] |
2506.02718 | Guanzhong Chen | Guanzhong Chen, Shaoxiong Yang, Chao Li, Wei Liu, Jian Luan, Zenglin Xu | Heterogeneous Group-Based Reinforcement Learning for LLM-based Multi-Agent Systems | null | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large Language Models (LLMs) have achieved remarkable success across diverse natural language processing tasks, yet their deployment in real-world applications is hindered by fixed knowledge cutoffs and difficulties in generating controllable, accurate outputs in a single inference. Multi-agent systems (MAS) built from specialized LLM agents offer a promising solution, enabling dynamic collaboration and iterative reasoning. However, optimizing these systems remains a challenge, as conventional methods such as prompt engineering and supervised fine-tuning entail high engineering overhead and limited adaptability. Reinforcement learning (RL), particularly multi-agent reinforcement learning (MARL), provides a scalable framework by refining agent policies based on system-level feedback. Nevertheless, existing MARL algorithms, such as Multi-Agent Proximal Policy Optimization (MAPPO), rely on Critic networks, which can cause training instability and increase computational burden. To address these limitations and target the prototypical Multi-Agent Search System (MASS), we propose Multi-Agent Heterogeneous Group Policy Optimization (MHGPO), a novel Critic-free algorithm that guides policy updates by estimating relative reward advantages across heterogeneous groups of rollouts. MHGPO eliminates the need for Critic networks, enhancing stability and reducing computational overhead. Additionally, we introduce three group rollout sampling strategies that trade off between efficiency and effectiveness. Experiments on a multi-agent LLM-based search system demonstrate that MHGPO consistently outperforms MAPPO in both task performance and computational efficiency, without requiring warm-up, underscoring its potential for stable and scalable optimization of complex LLM-based MAS. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:17:19 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Chen",
"Guanzhong",
""
],
[
"Yang",
"Shaoxiong",
""
],
[
"Li",
"Chao",
""
],
[
"Liu",
"Wei",
""
],
[
"Luan",
"Jian",
""
],
[
"Xu",
"Zenglin",
""
]
] |
2506.02720 | Xiaochong Lan | Xiaochong Lan, Jie Feng, Jiahuan Lei, Xinlei Shi, Yong Li | Benchmarking and Advancing Large Language Models for Local Life Services | KDD 2025 | null | null | null | cs.AI cs.CL | http://creativecommons.org/licenses/by/4.0/ | Large language models (LLMs) have exhibited remarkable capabilities and achieved significant breakthroughs across various domains, leading to their widespread adoption in recent years. Building on this progress, we investigate their potential in the realm of local life services. In this study, we establish a comprehensive benchmark and systematically evaluate the performance of diverse LLMs across a wide range of tasks relevant to local life services. To further enhance their effectiveness, we explore two key approaches: model fine-tuning and agent-based workflows. Our findings reveal that even a relatively compact 7B model can attain performance levels comparable to a much larger 72B model, effectively balancing inference cost and model capability. This optimization greatly enhances the feasibility and efficiency of deploying LLMs in real-world online services, making them more practical and accessible for local life applications. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:18:19 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Lan",
"Xiaochong",
""
],
[
"Feng",
"Jie",
""
],
[
"Lei",
"Jiahuan",
""
],
[
"Shi",
"Xinlei",
""
],
[
"Li",
"Yong",
""
]
] |
2506.02721 | Sean Raymond | Bernard Marty, Katherine R. Bermingham, Larry R. Nittler, Sean N. Raymond | Clues for Solar System Formation from Meteorites and their Parent Bodies | Chapter published in Comets III (pp. 681-730), edited by Karen J. Meech, Michael R. Combi, Dominique Bockelee-Morvan, Sean N. Raymond, and Michael E. Zolensky. University of Arizona Press, 2024. http://www.jstor.org/stable/jj.21819446.27 | null | null | null | astro-ph.EP astro-ph.SR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Understanding the origin of comets requires knowledge of how the Solar System formed from a cloud of dust and gas 4.567 Gyr ago. Here, a review is presented of how the remnants of this formation process, meteorites and to a lesser extent comets, shed light on Solar System evolution. The planets formed by a process of collisional agglomeration during the first hundred million years of Solar System history. The vast majority of the original population of planetary building blocks (~100 km-scale planetesimals) was either incorporated into the planets or removed from the system, via dynamical ejection or through a collision with the Sun. Only a small fraction of the original rocky planetesimals survive to this day in the form of asteroids (which represent a total of ~0.05% of Earth's mass) and comets. Meteorites are fragments of asteroids that have fallen to Earth, thereby providing scientists with samples of Solar System-scale processes for laboratory-based analysis. Meteorite datasets complement cometary datasets, which are predominantly obtained via remote observation as there are few cometary samples currently available for laboratory-based measurements. This chapter discusses how analysis of the mineralogical, elemental, and isotopic characteristics of meteorites provides insight into (i) the origin of matter that formed planets, (ii) the pressure, temperature, and chemical conditions that prevailed during planet formation, and (iii) a precise chronological framework of planetary accretion. Also examined is the use of stable isotope variations and nucleosynthetic isotope anomalies as constraints on the dynamics of the disk and planet formation, and how these data are integrated into new models of Solar System formation. It concludes with a discussion of Earth's accretion and its source of volatile elements, including water and organic species. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:18:30 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Marty",
"Bernard",
""
],
[
"Bermingham",
"Katherine R.",
""
],
[
"Nittler",
"Larry R.",
""
],
[
"Raymond",
"Sean N.",
""
]
] |
2506.02722 | Stephane Hess | Stephane Hess, David Bunch, Andrew Daly | Get me out of this hole: a profile likelihood approach to identifying and avoiding inferior local optima in choice models | null | null | null | null | econ.EM | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Choice modellers routinely acknowledge the risk of convergence to inferior local optima when using structures other than a simple linear-in-parameters logit model. At the same time, there is no consensus on appropriate mechanisms for addressing this issue. Most analysts seem to ignore the problem, while others try a set of different starting values, or put their faith in what they believe to be more robust estimation approaches. This paper puts forward the use of a profile likelihood approach that systematically analyses the parameter space around an initial maximum likelihood estimate and tests for the existence of better local optima in that space. We extend this to an iterative algorithm which then progressively searches for the best local optimum under given settings for the algorithm. Using a well known stated choice dataset, we show how the approach identifies better local optima for both latent class and mixed logit, with the potential for substantially different policy implications. In the case studies we conduct, an added benefit of the approach is that the new solutions exhibit properties that more closely adhere to the property of asymptotic normality, also highlighting the benefits of the approach in analysing the statistical properties of a solution. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:25:59 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Hess",
"Stephane",
""
],
[
"Bunch",
"David",
""
],
[
"Daly",
"Andrew",
""
]
] |
2506.02723 | Clemens S\"amann | Matteo Calisti, Christian Ketterer, Clemens S\"amann | Generalized cones admitting a curvature-dimension condition | 54 pages | null | null | null | math.DG gr-qc math-ph math.MG math.MP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We study (generalized) cones over metric spaces, both in Riemannian and Lorentzian signature. In particular, we establish synthetic lower Ricci curvature bounds \`a la Lott-Villani-Sturm and Ohta in the metric measure case, and \`a la Cavalletti-Mondino in Lorentzian signature. Here, a generalized cone is a warped product of a one-dimensional base space, which will be positive or negative definite, over a fiber that is a metric space. We prove that Riemannian or Lorentzian generalized cones over $\mathsf{CD}$-spaces satisfy the (timelike) measure contraction property $\mathsf{(T)MCP}$ - a weaker version of a (timelike) curvature-dimension condition $\mathsf{(T)CD}$. Conversely, if the generalized cone is a $\mathsf{(T)CD}$-space, then the fiber is a $\mathsf{CD}$-space with the appropriate bounds on Ricci curvature and dimension. In proving these results we develop a novel and powerful two-dimensional localization technique, which we expect to be interesting in its own right and useful in other circumstances. We conclude by giving several applications including synthetic singularity and splitting theorems for generalized cones. The final application is that we propose a new definition for lower curvature bounds for metric and metric measure spaces via lower curvature bounds for generalized cones over the given space. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:28:53 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Calisti",
"Matteo",
""
],
[
"Ketterer",
"Christian",
""
],
[
"Sämann",
"Clemens",
""
]
] |
2506.02724 | Andrey Veprikov | Andrey Veprikov, Vladimir Solodkin, Alexander Zyl, Andrey Savchenko and Aleksandr Beznosikov | WeightLoRA: Keep Only Necessary Adapters | 13 pages, 9 tables | null | null | null | cs.LG math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The widespread utilization of language models in modern applications is inconceivable without Parameter-Efficient Fine-Tuning techniques, such as low-rank adaptation ($\texttt{LoRA}$), which adds trainable adapters to selected layers. Although $\texttt{LoRA}$ may obtain accurate solutions, it requires significant memory to train large models and intuition on which layers to add adapters. In this paper, we propose a novel method, $\texttt{WeightLoRA}$, which overcomes this issue by adaptive selection of the most critical $\texttt{LoRA}$ heads throughout the optimization process. As a result, we can significantly reduce the number of trainable parameters while maintaining the capability to obtain consistent or even superior metric values. We conduct experiments for a series of competitive benchmarks and DeBERTa, BART, and Llama models, comparing our method with different adaptive approaches. The experimental results demonstrate the efficacy of $\texttt{WeightLoRA}$ and the superior performance of $\texttt{WeightLoRA+}$ in almost all cases. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:33:16 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Veprikov",
"Andrey",
""
],
[
"Solodkin",
"Vladimir",
""
],
[
"Zyl",
"Alexander",
""
],
[
"Savchenko",
"Andrey",
""
],
[
"Beznosikov",
"Aleksandr",
""
]
] |
2506.02725 | Jie Huang | Jie Huang, Fanlin Jia, Xiao He | Recursive Privacy-Preserving Estimation Over Markov Fading Channels | 12 pages, 5 figures | null | null | null | eess.SY cs.SY | http://creativecommons.org/licenses/by/4.0/ | In industrial applications, the presence of moving machinery, vehicles, and personnel, contributes to the dynamic nature of the wireless channel. This time variability induces channel fading, which can be effectively modeled using a Markov fading channel (MFC). In this paper, we investigate the problem of secure state estimation for systems that communicate over a MFC in the presence of an eavesdropper. The objective is to enable a remote authorized user to accurately estimate the states of a dynamic system, while considering the potential interception of the sensor's packet through a wiretap channel. To prevent information leakage, a novel co-design strategy is established, which combines a privacy-preserving mechanism with a state estimator. To implement our encoding scheme, a nonlinear mapping of the innovation is introduced based on the weighted reconstructed innovation previously received by the legitimate user. Corresponding to this encoding scheme, we design a recursive privacy-preserving filtering algorithm to achieve accurate estimation. The boundedness of estimation error dynamics at the legitimate user's side is discussed and the divergence of the eavesdropper's estimation error is analyzed, which demonstrates the effectiveness of our co-design strategy in ensuring secrecy. Furthermore, a simulation example of a three-tank system is provided to demonstrate the effectiveness and feasibility of our privacy-preserving estimation method. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:33:49 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Huang",
"Jie",
""
],
[
"Jia",
"Fanlin",
""
],
[
"He",
"Xiao",
""
]
] |
2506.02726 | Feifan Liu | Qihang Yan, Xinyu Zhang, Luming Guo, Qi Zhang, Feifan Liu | RACE-Align: Retrieval-Augmented and Chain-of-Thought Enhanced Preference Alignment for Large Language Models | null | null | null | null | cs.CL cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large Language Models (LLMs) struggle with accuracy, domain-specific reasoning, and interpretability in vertical domains. Traditional preference alignment methods like Reinforcement Learning from Human Feedback (RLHF) and Direct Preference Optimization (DPO) often overlook the underlying knowledge sources and reasoning logic. This paper introduces RACE-Align (Retrieval-Augmented and Chain-of-Thought Enhanced Alignment), a novel framework designed to address these limitations. RACE-Align systematically constructs a binary preference dataset incorporating external knowledge support and explicit Chain-of-Thought (CoT) reasoning, then aligns LLMs using the DPO algorithm. The core innovation lies in its preference data construction strategy: it integrates AI-driven retrieval for factual grounding, enhancing knowledgeability and accuracy, and emphasizes the optimization of domain-specific CoT, treating the reasoning process itself as a key preference dimension. A multi-stage, AI-driven refinement pipeline cost-effectively generates these preference pairs. Experimental validation in Traditional Chinese Medicine (TCM) using Qwen3-1.7B as the base model demonstrates that RACE-Align significantly outperforms the original base model and a model fine-tuned only with Supervised Fine-Tuning (SFT). Improvements were observed across multiple dimensions, including answer accuracy, information richness, application of TCM thinking patterns, logicality and depth of reasoning, and interpretability. These findings suggest RACE-Align offers an effective pathway to enhance LLMs' knowledge application, reasoning reliability, and process transparency in complex vertical domains. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:36:38 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Yan",
"Qihang",
""
],
[
"Zhang",
"Xinyu",
""
],
[
"Guo",
"Luming",
""
],
[
"Zhang",
"Qi",
""
],
[
"Liu",
"Feifan",
""
]
] |
2506.02727 | Christian Gang Liu | Christian Gang Liu, Peter Bodorik, Dawn Jutla | Transforming Automatically BPMN Models to Smart Contracts with Nested Collaborative Transactions (TABS+) | Preprint. arXiv admin note: substantial text overlap with arXiv:2505.24309 | Distributed Ledger Technologies: Research and Practice, Volume 3, Issue 3 Article No.: 21, Pages 1 - 37, 2024 | 10.1145/3654802 | null | cs.SE | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Development of blockchain smart contracts is more difficult than mainstream software development because the underlying blockchain infrastructure poses additional complexity. To ease the developer's task of writing smart contract, as other research efforts, we also use Business Process Model and Notation BPMN modeling to describe application requirements for trade of goods and services and then transform automatically the BPMN model into the methods of a smart contract. In our previous research we described our approach and a tool to Transform Automatically BPMN models into Smart contracts TABS. In this paper, we describe how the TABS approach is augmented with the support for a BPMN collaborative transaction by several actors. Our approach analyzes the BPMN model to determine which patterns in the BPMN model are suitable for use as collaborative transactions. The found BPMN patterns that are suitable as transactions are shown to the developer who decides which ones should be deployed as collaborative transactions. We describe how our approach automatically transform the BPMN model into smart contract the provides a transaction mechanism to enforce the transactional properties of the nested transactions. Our approach greatly reduces the developers task as synchronization of collaborative activities is provided by our approach, so that the developer needs to code only independent tasks with well-defined inputs and outputs. We also overview the TABS+ tool we built as a proof of concept to show that our approach is feasible. Finally, we provide estimates on the cost of supporting the nested BPMN collaborative transactions. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:37:41 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Liu",
"Christian Gang",
""
],
[
"Bodorik",
"Peter",
""
],
[
"Jutla",
"Dawn",
""
]
] |
2506.02728 | Michael Brandenbursky | Michael Brandenbursky and Lior Menashe | Bounded cohomology of measure-preserving homeomorphism groups of non-orientable surfaces | 14 pages, 5 figures | null | null | null | math.GT math.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Let $N_g$ be a closed non-orientable surface of genus $g\geq 3$. Let $\operatorname{Homeo}_0(N_g,\mu)$ be the identity component of the group of measure-preserving homeomorphisms of $N_g$. In this work we prove that the third bounded cohomology of $\operatorname{Homeo}_0(N_g,\mu)$ is infinite dimensional. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:38:38 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Brandenbursky",
"Michael",
""
],
[
"Menashe",
"Lior",
""
]
] |
2506.02729 | Salih Kibaro\u{g}lu | Salih Kibaro\u{g}lu and Mustafa Senay | Generalized Brans-Dicke theory from Verlinde's entropic gravity | 14 pages | Nucl. Phys. B 1017 (2025) 116974 | 10.1016/j.nuclphysb.2025.116974 | null | gr-qc | http://creativecommons.org/licenses/by/4.0/ | In this work, we develop a q-deformed scalar-tensor theory of gravitation by combining Verlinde's entropic gravity paradigm with statistical deformation effects. The resulting model modifies the Brans-Dicke framework through a deformation function, treated as a constant rescaling factor for the effective gravitational coupling. We derive the corresponding q-deformed field equations and analyze their theoretical consistency including the recovery of standard gravitational models in specific limits. While the present formulation preserves key symmetries and provides a generalized description of gravitational dynamics, it does not yield a significant deviation from Brans-Dicke theory. The study concludes with prospects for future research, including the exploration of cosmological solutions arising from the q-deformed field equations. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:42:30 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Kibaroğlu",
"Salih",
""
],
[
"Senay",
"Mustafa",
""
]
] |
2506.02730 | Po-Chieh Yu | Po-Chieh Yu | An Exploratory Framework for Future SETI Applications: Detecting Generative Reactivity via Language Models | submitted to the International Journal of Astrobiology | null | null | null | astro-ph.IM cs.CL | http://creativecommons.org/licenses/by-nc-sa/4.0/ | We present an exploratory framework to test whether noise-like input can induce structured responses in language models. Instead of assuming that extraterrestrial signals must be decoded, we evaluate whether inputs can trigger linguistic behavior in generative systems. This shifts the focus from decoding to viewing structured output as a sign of underlying regularity in the input. We tested GPT-2 small, a 117M-parameter model trained on English text, using four types of acoustic input: human speech, humpback whale vocalizations, Phylloscopus trochilus birdsong, and algorithmically generated white noise. All inputs were treated as noise-like, without any assumed symbolic encoding. To assess reactivity, we defined a composite score called Semantic Induction Potential (SIP), combining entropy, syntax coherence, compression gain, and repetition penalty. Results showed that whale and bird vocalizations had higher SIP scores than white noise, while human speech triggered only moderate responses. This suggests that language models may detect latent structure even in data without conventional semantics. We propose that this approach could complement traditional SETI methods, especially in cases where communicative intent is unknown. Generative reactivity may offer a different way to identify data worth closer attention. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:46:57 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Yu",
"Po-Chieh",
""
]
] |
2506.02732 | Markus Johannes Stroppel | Markus Johannes Stroppel | Construction of the smallest Ree-Tits unital from the special linear group of degree two over the field with eight elements | null | null | null | null | math.GR math.CO | http://creativecommons.org/licenses/by-nc-sa/4.0/ | We construct the smallest Ree-Tits unital from a group of matrices that is isomorphic to the commutator group of the corresponding Ree group. The matrix description is used to determine configurations in the unital via explicit computations. Embeddings into larger Ree-Tits unitals are made explicit. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:50:04 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Stroppel",
"Markus Johannes",
""
]
] |
2506.02733 | Dagang Li | Xiaoyi Feng, Kaifeng Zou, Caichun Cen, Tao Huang, Hui Guo, Zizhou Huang, Yingli Zhao, Mingqing Zhang, Diwei Wang, Yuntao Zou, Dagang Li | LinkTo-Anime: A 2D Animation Optical Flow Dataset from 3D Model Rendering | null | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Existing optical flow datasets focus primarily on real-world simulation or synthetic human motion, but few are tailored to Celluloid(cel) anime character motion: a domain with unique visual and motion characteristics. To bridge this gap and facilitate research in optical flow estimation and downstream tasks such as anime video generation and line drawing colorization, we introduce LinkTo-Anime, the first high-quality dataset specifically designed for cel anime character motion generated with 3D model rendering. LinkTo-Anime provides rich annotations including forward and backward optical flow, occlusion masks, and Mixamo Skeleton. The dataset comprises 395 video sequences, totally 24,230 training frames, 720 validation frames, and 4,320 test frames. Furthermore, a comprehensive benchmark is constructed with various optical flow estimation methods to analyze the shortcomings and limitations across multiple datasets. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:50:20 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Feng",
"Xiaoyi",
""
],
[
"Zou",
"Kaifeng",
""
],
[
"Cen",
"Caichun",
""
],
[
"Huang",
"Tao",
""
],
[
"Guo",
"Hui",
""
],
[
"Huang",
"Zizhou",
""
],
[
"Zhao",
"Yingli",
""
],
[
"Zhang",
"Mingqing",
""
],
[
"Wang",
"Diwei",
""
],
[
"Zou",
"Yuntao",
""
],
[
"Li",
"Dagang",
""
]
] |
2506.02734 | Wei-Wei Zhang | Wei-Wei Zhang, Zhuo Xia, Wei Zhao, Wei Pan, Haobin Shi | Self-attention U-Net decoder for toric codes | 16 pages; 12 figures; | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In the NISQ era, one of the most important bottlenecks for the realization of universal quantum computation is error correction. Stabiliser code is the most recognizable type of quantum error correction code. A scalable efficient decoder is most desired for the application of the quantum error correction codes. In this work, we propose a self-attention U-Net quantum decoder (SU-NetQD) for toric code, which outperforms the minimum weight perfect matching decoder, especially in the circuit level noise environments. Specifically, with our SU-NetQD, we achieve lower logical error rates compared with MWPM and discover an increased trend of code threshold as the increase of noise bias. We obtain a high threshold of 0.231 for the extremely biased noise environment. The combination of low-level decoder and high-level decoder is the key innovation for the high accuracy of our decoder. With transfer learning mechanics, our decoder is scalable for cases with different code distances. Our decoder provides a practical tool for quantum noise analysis and promotes the practicality of quantum error correction codes and quantum computing. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:50:47 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Zhang",
"Wei-Wei",
""
],
[
"Xia",
"Zhuo",
""
],
[
"Zhao",
"Wei",
""
],
[
"Pan",
"Wei",
""
],
[
"Shi",
"Haobin",
""
]
] |
2506.02735 | Min Fu | Min Fu, Lipeng Zhu, and Rui Zhang | Extremely Large-Scale Movable Antenna-Enabled Multiuser Communications: Modeling and Optimization | 13 pages | null | null | null | cs.IT eess.SP math.IT | http://creativecommons.org/licenses/by/4.0/ | Movable antenna (MA) has been recognized as a promising technology to improve communication performance in future wireless networks such as 6G. To unleash its potential, this paper proposes a novel architecture, namely extremely large-scale MA (XL-MA), which allows flexible antenna/subarray positioning over an extremely large spatial region for effectively enhancing near-field effects and spatial multiplexing performance. In particular, this paper studies an uplink XL-MA-enabled multiuser system, where single-antenna users distributed in a coverage area are served by a base station (BS) equipped with multiple movable subarrays. We begin by presenting a spatially non-stationary channel model to capture the near-field effects, including positiondependent large-scale channel gains and line-of-sight visibility. To evaluate system performance, we further derive a closedform approximation of the expected weighted sum rate under maximum ratio combining (MRC), revealing that optimizing XLMA placement enhances user channel power gain to increase desired signal power and reduces channel correlation to decreases multiuser interference. Building upon this, we formulate an antenna placement optimization problem to maximize the expected weighted sum rate, leveraging statistical channel conditions and user distribution. To efficiently solve this challenging non-linear binary optimization problem, we propose a polynomial-time successive replacement algorithm. Simulation results demonstrate that the proposed XL-MA placement strategy achieves nearoptimal performance, significantly outperforming benchmark schemes based on conventional fixed-position antennas. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:50:51 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Fu",
"Min",
""
],
[
"Zhu",
"Lipeng",
""
],
[
"Zhang",
"Rui",
""
]
] |
2506.02736 | Shufan Qing | Shufan Qing, Anzhen Li, Qiandi Wang, Yuefeng Niu, Mingchen Feng, Guoliang Hu, Jinqiao Wu, Fengtao Nan, Yingchun Fan | GeneA-SLAM2: Dynamic SLAM with AutoEncoder-Preprocessed Genetic Keypoints Resampling and Depth Variance-Guided Dynamic Region Removal | null | null | null | null | cs.CV cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Existing semantic SLAM in dynamic environments mainly identify dynamic regions through object detection or semantic segmentation methods. However, in certain highly dynamic scenarios, the detection boxes or segmentation masks cannot fully cover dynamic regions. Therefore, this paper proposes a robust and efficient GeneA-SLAM2 system that leverages depth variance constraints to handle dynamic scenes. Our method extracts dynamic pixels via depth variance and creates precise depth masks to guide the removal of dynamic objects. Simultaneously, an autoencoder is used to reconstruct keypoints, improving the genetic resampling keypoint algorithm to obtain more uniformly distributed keypoints and enhance the accuracy of pose estimation. Our system was evaluated on multiple highly dynamic sequences. The results demonstrate that GeneA-SLAM2 maintains high accuracy in dynamic scenes compared to current methods. Code is available at: https://github.com/qingshufan/GeneA-SLAM2. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:51:53 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Qing",
"Shufan",
""
],
[
"Li",
"Anzhen",
""
],
[
"Wang",
"Qiandi",
""
],
[
"Niu",
"Yuefeng",
""
],
[
"Feng",
"Mingchen",
""
],
[
"Hu",
"Guoliang",
""
],
[
"Wu",
"Jinqiao",
""
],
[
"Nan",
"Fengtao",
""
],
[
"Fan",
"Yingchun",
""
]
] |
2506.02737 | James Urquhart | A. L. Patel (1), J. S. Urquhart (1), A. Y. Yang (2,3), L. K. Morgan (4), K. M. Menten (5), M. A. Thompson (6), T. Moore (7), I. Grozdanova (1), S. Khan (5), T. Csengeri (8) ((1) Kent, (2) NAOC, (3) Key Laboratory of Radio Astronomy and Technology, Chinese Academy of Sciences, (4) GBT, (5) LJMU, (6) MPIfR, (7) Leeds, (8) Univ. Bordeaux) | SCOTCH III: Complete search for Hypercompact HII regions in the fourth quadrant | 18 pages, 15 figures. Published in MNRAS Vol 538, Issue 4, April 2025 | null | 10.1093/mnras/staf450 | null | astro-ph.GA | http://creativecommons.org/licenses/by/4.0/ | We present high-frequency (18-24 GHz) radio continuum observations towards 335 methanol masers, excellent signposts for young, embedded high-mass protostars. These complete the search for hypercompact HII (HCHII) regions towards young high-mass star-forming clumps within the fourth quadrant of the Galactic plane. HCHII regions are the earliest observable signatures of radio continuum emission from high-mass stars ionizing their surroundings, though their rarity and short lifetimes make them challenging to study. We have observed methanol maser sites at 20-arcsec resolution and identified 121 discrete high-frequency radio sources. Of these, 42 compact sources are embedded in dense clumps and coincide with methanol masers, making them as excellent HCHII region candidates. These sources were followed up at higher resolution (0.5-arcsec) for confirmation. We constructed spectral energy distributions across 5-24 GHz to determine their physical properties, fitting either a simple HII region model or a power-law as needed. This analysis identified 20 HCHII regions, 9 intermediate objects, 3 UCHII regions, and 3 radio jet candidates. Combining these results with previous findings, the SCOTCH survey has identified 33 HCHII regions, 15 intermediate objects, 9 UCHII regions, and 4 radio jet candidates, tripling the known number of HCHII regions. Eleven of these sources remain optically thick at 24 GHz. This survey provides a valuable sample of the youngest HII regions and insights into early massive star formation. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:51:58 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Patel",
"A. L.",
"",
"Kent"
],
[
"Urquhart",
"J. S.",
"",
"Kent"
],
[
"Yang",
"A. Y.",
"",
"NAOC",
"Key Laboratory of Radio Astronomy and Technology, Chinese Academy of Sciences"
],
[
"Morgan",
"L. K.",
"",
"GBT"
],
[
"Menten",
"K. M.",
"",
"LJMU"
],
[
"Thompson",
"M. A.",
"",
"MPIfR"
],
[
"Moore",
"T.",
"",
"Leeds"
],
[
"Grozdanova",
"I.",
"",
"Kent"
],
[
"Khan",
"S.",
"",
"LJMU"
],
[
"Csengeri",
"T.",
"",
"Univ. Bordeaux"
]
] |
2506.02739 | Pengcheng Zhou | Pengcheng Zhou, Yinglun Feng, Halimulati Julaiti, Zhongliang Yang | Why do AI agents communicate in human language? | null | null | null | null | cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large Language Models (LLMs) have become foundational to modern AI agent systems, enabling autonomous agents to reason and plan. In most existing systems, inter-agent communication relies primarily on natural language. While this design supports interpretability and human oversight, we argue that it introduces fundamental limitations in agent-to-agent coordination. The semantic space of natural language is structurally misaligned with the high-dimensional vector spaces in which LLMs operate, resulting in information loss and behavioral drift. Beyond surface-level inefficiencies, we highlight a deeper architectural limitation: current LLMs were not trained with the objective of supporting agentic behavior. As such, they lack mechanisms for modeling role continuity, task boundaries, and multi-agent dependencies. The standard next-token prediction paradigm fails to support the structural alignment required for robust, scalable agent coordination. Based on this, we argue that two core questions deserve careful examination: first, given that AI agents fundamentally operate in high-dimensional vector spaces, should they rely on a language system originally designed for human cognition as their communication medium? Second, should we consider developing a new model construction paradigm that builds models from the ground up to natively support structured communication, shared intentionality, and task alignment in multi-role, multi-agent environments? This paper calls for a reconsideration not only of how agents should communicate, but also of what it fundamentally means to train a model that natively supports multi-agent coordination and communication. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:53:29 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Zhou",
"Pengcheng",
""
],
[
"Feng",
"Yinglun",
""
],
[
"Julaiti",
"Halimulati",
""
],
[
"Yang",
"Zhongliang",
""
]
] |
2506.02740 | Ama\c{c} Herda\u{g}delen | Ama\c{c} Herda\u{g}delen and Marco Baroni | Stereotypical gender actions can be extracted from Web text | null | Herda\u{g}delen, Ama\c{c}, and Marco Baroni. "Stereotypical gender actions can be extracted from web text." Journal of the American Society for Information Science and Technology 62.9 (2011): 1741-1749 | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | We extracted gender-specific actions from text corpora and Twitter, and compared them to stereotypical expectations of people. We used Open Mind Common Sense (OMCS), a commonsense knowledge repository, to focus on actions that are pertinent to common sense and daily life of humans. We use the gender information of Twitter users and Web-corpus-based pronoun/name gender heuristics to compute the gender bias of the actions. With high recall, we obtained a Spearman correlation of 0.47 between corpus-based predictions and a human gold standard, and an area under the ROC curve of 0.76 when predicting the polarity of the gold standard. We conclude that it is feasible to use natural text (and a Twitter-derived corpus in particular) in order to augment commonsense repositories with the stereotypical gender expectations of actions. We also present a dataset of 441 commonsense actions with human judges' ratings on whether the action is typically/slightly masculine/feminine (or neutral), and another larger dataset of 21,442 actions automatically rated by the methods we investigate in this study. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:55:00 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Herdağdelen",
"Amaç",
""
],
[
"Baroni",
"Marco",
""
]
] |
2506.02741 | Pengchong Hu | Pengchong Hu, Zhizhong Han | VTGaussian-SLAM: RGBD SLAM for Large Scale Scenes with Splatting View-Tied 3D Gaussians | ICML 2025 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Jointly estimating camera poses and mapping scenes from RGBD images is a fundamental task in simultaneous localization and mapping (SLAM). State-of-the-art methods employ 3D Gaussians to represent a scene, and render these Gaussians through splatting for higher efficiency and better rendering. However, these methods cannot scale up to extremely large scenes, due to the inefficient tracking and mapping strategies that need to optimize all 3D Gaussians in the limited GPU memories throughout the training to maintain the geometry and color consistency to previous RGBD observations. To resolve this issue, we propose novel tracking and mapping strategies to work with a novel 3D representation, dubbed view-tied 3D Gaussians, for RGBD SLAM systems. View-tied 3D Gaussians is a kind of simplified Gaussians, which is tied to depth pixels, without needing to learn locations, rotations, and multi-dimensional variances. Tying Gaussians to views not only significantly saves storage but also allows us to employ many more Gaussians to represent local details in the limited GPU memory. Moreover, our strategies remove the need of maintaining all Gaussians learnable throughout the training, while improving rendering quality, and tracking accuracy. We justify the effectiveness of these designs, and report better performance over the latest methods on the widely used benchmarks in terms of rendering and tracking accuracy and scalability. Please see our project page for code and videos at https://machineperceptionlab.github.io/VTGaussian-SLAM-Project . | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:59:19 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Hu",
"Pengchong",
""
],
[
"Han",
"Zhizhong",
""
]
] |
2506.02742 | Xiaoxue Gao | Xiaoxue Gao, Huayun Zhang and Nancy F. Chen | Prompt-Unseen-Emotion: Zero-shot Expressive Speech Synthesis with Prompt-LLM Contextual Knowledge for Mixed Emotions | null | null | null | null | eess.AS cs.AI cs.SD eess.SP | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Existing expressive text-to-speech (TTS) systems primarily model a limited set of categorical emotions, whereas human conversations extend far beyond these predefined emotions, making it essential to explore more diverse emotional speech generation for more natural interactions. To bridge this gap, this paper proposes a novel prompt-unseen-emotion (PUE) approach to generate unseen emotional speech via emotion-guided prompt learning. PUE is trained utilizing an LLM-TTS architecture to ensure emotional consistency between categorical emotion-relevant prompts and emotional speech, allowing the model to quantitatively capture different emotion weightings per utterance. During inference, mixed emotional speech can be generated by flexibly adjusting emotion proportions and leveraging LLM contextual knowledge, enabling the model to quantify different emotional styles. Our proposed PUE successfully facilitates expressive speech synthesis of unseen emotions in a zero-shot setting. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 10:59:22 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Gao",
"Xiaoxue",
""
],
[
"Zhang",
"Huayun",
""
],
[
"Chen",
"Nancy F.",
""
]
] |
2506.02743 | Thi Nhung Dao | Thi Nhung Dao, Martin Gabelmann, Margarete M\"uhlleitner | Loop-corrected Trilinear Higgs Self-Couplings in the NMSSM with Inverse Seesaw Mechanism | 34 pages, 5 figures, 3 tables | null | null | KA-TP-16-2025, FR-PHENO-2025-006 | hep-ph | http://creativecommons.org/licenses/by/4.0/ | The higher-order corrections for the SM-like Higgs boson mass and the trilinear Higgs self-couplings in the Next-to-Minimal Supersymmetric extension of the Standard Model (NMSSM) with Inverse Seesaw Mechanism are significant and highly correlated. We present here the full one-loop corrections to the trilinear Higgs self-couplings supplemented by the dominant top-Yukawa and strong coupling induced two-loop corrections from our previous calculations in the complex NMSSM. These corrections are performed consistently with the corresponding Higgs boson mass corrections. We discuss in detail the new effects from the extended neutrino and sneutrino sectors on both the trilinear Higgs self-couplings and the SM-like Higgs boson mass. When compared to the case of the NMSSM without Inverse Seesaw Mechanism, the new effects can be up to 10\% for the effective SM-like trilinear Higgs self-couplings, and up to 4.5\% for the SM-like Higgs boson mass for valid parameter points, i.e. points satisfying the Higgs data, the neutrino data, the constraints from the charged lepton flavor-violating decays, and the new physics constraints from the oblique parameters $S, T, U$. The new corrections are also included in the Higgs-to-Higgs decays for the heavy Higgs states and implemented in the new version of the Fortran code NMSSMCALC-nuSS. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:00:03 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Dao",
"Thi Nhung",
""
],
[
"Gabelmann",
"Martin",
""
],
[
"Mühlleitner",
"Margarete",
""
]
] |
2506.02744 | Junyuan Liu | Junyuan Liu, Xinglei Wang, and Tao Cheng | Enriching Location Representation with Detailed Semantic Information | null | null | null | null | cs.CE cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Spatial representations that capture both structural and semantic characteristics of urban environments are essential for urban modeling. Traditional spatial embeddings often prioritize spatial proximity while underutilizing fine-grained contextual information from places. To address this limitation, we introduce CaLLiPer+, an extension of the CaLLiPer model that systematically integrates Point-of-Interest (POI) names alongside categorical labels within a multimodal contrastive learning framework. We evaluate its effectiveness on two downstream tasks, land use classification and socioeconomic status distribution mapping, demonstrating consistent performance gains of 4% to 11% over baseline methods. Additionally, we show that incorporating POI names enhances location retrieval, enabling models to capture complex urban concepts with greater precision. Ablation studies further reveal the complementary role of POI names and the advantages of leveraging pretrained text encoders for spatial representations. Overall, our findings highlight the potential of integrating fine-grained semantic attributes and multimodal learning techniques to advance the development of urban foundation models. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:06:51 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Liu",
"Junyuan",
""
],
[
"Wang",
"Xinglei",
""
],
[
"Cheng",
"Tao",
""
]
] |
2506.02745 | Maelie Mondelin | M. Mondelin, F. Bournaud, J.-C. Cuillandre, S. Codis, C. Stone, M. Bolzonella, J.G. Sorce, M. Kluge, N.A. Hatch, F.R. Marleau, M. Schirmer, H. Bouy, F. Buitrago, C. Tortora, L. Quilley, K. George, M. Baes, T. Saifollahi, P.M. Sanchez-Alarcon, J.H. Knapen, N. Aghanim, A. Amara, S. Andreon, C. Baccigalupi, A. Balestra, S. Bardelli, P. Battaglia, A. Biviano, E. Branchini, M. Brescia, J. Brinchmann, V. Capobianco, C. Carbone, M. Castellano, G. Castignani, S. Cavuoti, A. Cimatti, G. Congedo, C.J. Conselice, L. Conversi, Y. Copin, F. Courbin, H.M. Courtois, M. Cropper, G. De Lucia, X. Dupac, M. Fabricius, M. Farina, F. Faustini, S. Ferriol, S. Fotopoulou, B. Gillis, C. Giocoli, F. Grupp, S.V.H. Haugan, W. Holmes, F. Hormuth, A. Hornstrup, K. Jahnke, M. Jhabvala, E. Keih"anen, S. Kermiche, M. Kilbinger, B. Kubik, M. K"ummel, H. Kurki-Suonio, A.M.C. Le Brun, S. Ligori, P.B. Lilje, V. Lindholm, I. Lloro, D. Maino, E. Maiorano, O. Mansutti, S. Marcin, O. Marggraf, M. Martinelli, E. Medinaceli, Y. Mellier, E. Merlin, G. Meylan, L. Moscardini, S.-M. Niemi, C. Padilla, F. Pasian, K. Pedersen, W.J. Percival, V. Pettorino, S. Pires, M. Poncet, L.A. Popa, L. Pozzetti, A. Renzi, G. Riccio, E. Romelli, R. Saglia, P. Schneider, A. Secroun, S. Serrano, C. Sirignano, J. Steinwagner, I. Tereno, R. Toledo-Moreo, F. Torradeflot, I. Tutusaus, L. Valenziano, T. Vassallo, Y. Wang, J. Weller, F.M. Zerbi, E. Zucca, C. Burigana, V. Scottez | Euclid: Early Release Observations -- The surface brightness and colour profiles of the far outskirts of galaxies in the Perseus cluster | 29 pages, 20 figures | null | 10.1051/0004-6361/202554838 | null | astro-ph.GA | http://creativecommons.org/licenses/by/4.0/ | The Perseus field captured by Euclid as part of its Early Release Observations provides a unique opportunity to study cluster environment ranging from outskirts to dense regions. Leveraging unprecedented optical and near-infrared depths, we investigate the stellar structure of massive disc galaxies in this field. This study focuses on outer disc profiles, including simple exponential (Type I), down- (Type II) and up-bending break (Type III) profiles, and their associated colour gradients, to trace late assembly processes across various environments. Type II profiles, though relatively rare in high dense environments, appear stabilised by internal mechanisms like bars and resonances, even within dense cluster cores. Simulations suggest that in dense environments, Type II profiles tend to evolve into Type I profiles over time. Type III profiles often exhibit small colour gradients beyond the break, hinting at older stellar populations, potentially due to radial migration or accretion events. We analyse correlations between galaxy mass, morphology, and profile types. Mass distributions show weak trends of decreasing mass from the centre to the outskirts of the Perseus cluster. Type III profiles become more prevalent, while Type I profiles decrease in lower-mass galaxies with cluster centric distance. Type I profiles dominate in spiral galaxies, while Type III profiles are more common in S0 galaxies. Type II profiles are consistently observed across all morphological types. While the limited sample size restricts statistical power, our findings shed light on the mechanisms shaping galaxy profiles in cluster environments. Future work should extend observations to the cluster outskirts to enhance statistical significance. Additionally, 3D velocity maps are needed to achieve a non-projected view of galaxy positions, offering deeper insights into spatial distribution and dynamics. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:07:15 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Mondelin",
"M.",
""
],
[
"Bournaud",
"F.",
""
],
[
"Cuillandre",
"J. -C.",
""
],
[
"Codis",
"S.",
""
],
[
"Stone",
"C.",
""
],
[
"Bolzonella",
"M.",
""
],
[
"Sorce",
"J. G.",
""
],
[
"Kluge",
"M.",
""
],
[
"Hatch",
"N. A.",
""
],
[
"Marleau",
"F. R.",
""
],
[
"Schirmer",
"M.",
""
],
[
"Bouy",
"H.",
""
],
[
"Buitrago",
"F.",
""
],
[
"Tortora",
"C.",
""
],
[
"Quilley",
"L.",
""
],
[
"George",
"K.",
""
],
[
"Baes",
"M.",
""
],
[
"Saifollahi",
"T.",
""
],
[
"Sanchez-Alarcon",
"P. M.",
""
],
[
"Knapen",
"J. H.",
""
],
[
"Aghanim",
"N.",
""
],
[
"Amara",
"A.",
""
],
[
"Andreon",
"S.",
""
],
[
"Baccigalupi",
"C.",
""
],
[
"Balestra",
"A.",
""
],
[
"Bardelli",
"S.",
""
],
[
"Battaglia",
"P.",
""
],
[
"Biviano",
"A.",
""
],
[
"Branchini",
"E.",
""
],
[
"Brescia",
"M.",
""
],
[
"Brinchmann",
"J.",
""
],
[
"Capobianco",
"V.",
""
],
[
"Carbone",
"C.",
""
],
[
"Castellano",
"M.",
""
],
[
"Castignani",
"G.",
""
],
[
"Cavuoti",
"S.",
""
],
[
"Cimatti",
"A.",
""
],
[
"Congedo",
"G.",
""
],
[
"Conselice",
"C. J.",
""
],
[
"Conversi",
"L.",
""
],
[
"Copin",
"Y.",
""
],
[
"Courbin",
"F.",
""
],
[
"Courtois",
"H. M.",
""
],
[
"Cropper",
"M.",
""
],
[
"De Lucia",
"G.",
""
],
[
"Dupac",
"X.",
""
],
[
"Fabricius",
"M.",
""
],
[
"Farina",
"M.",
""
],
[
"Faustini",
"F.",
""
],
[
"Ferriol",
"S.",
""
],
[
"Fotopoulou",
"S.",
""
],
[
"Gillis",
"B.",
""
],
[
"Giocoli",
"C.",
""
],
[
"Grupp",
"F.",
""
],
[
"Haugan",
"S. V. H.",
""
],
[
"Holmes",
"W.",
""
],
[
"Hormuth",
"F.",
""
],
[
"Hornstrup",
"A.",
""
],
[
"Jahnke",
"K.",
""
],
[
"Jhabvala",
"M.",
""
],
[
"Keih\"anen",
"E.",
""
],
[
"Kermiche",
"S.",
""
],
[
"Kilbinger",
"M.",
""
],
[
"Kubik",
"B.",
""
],
[
"K\"ummel",
"M.",
""
],
[
"Kurki-Suonio",
"H.",
""
],
[
"Brun",
"A. M. C. Le",
""
],
[
"Ligori",
"S.",
""
],
[
"Lilje",
"P. B.",
""
],
[
"Lindholm",
"V.",
""
],
[
"Lloro",
"I.",
""
],
[
"Maino",
"D.",
""
],
[
"Maiorano",
"E.",
""
],
[
"Mansutti",
"O.",
""
],
[
"Marcin",
"S.",
""
],
[
"Marggraf",
"O.",
""
],
[
"Martinelli",
"M.",
""
],
[
"Medinaceli",
"E.",
""
],
[
"Mellier",
"Y.",
""
],
[
"Merlin",
"E.",
""
],
[
"Meylan",
"G.",
""
],
[
"Moscardini",
"L.",
""
],
[
"Niemi",
"S. -M.",
""
],
[
"Padilla",
"C.",
""
],
[
"Pasian",
"F.",
""
],
[
"Pedersen",
"K.",
""
],
[
"Percival",
"W. J.",
""
],
[
"Pettorino",
"V.",
""
],
[
"Pires",
"S.",
""
],
[
"Poncet",
"M.",
""
],
[
"Popa",
"L. A.",
""
],
[
"Pozzetti",
"L.",
""
],
[
"Renzi",
"A.",
""
],
[
"Riccio",
"G.",
""
],
[
"Romelli",
"E.",
""
],
[
"Saglia",
"R.",
""
],
[
"Schneider",
"P.",
""
],
[
"Secroun",
"A.",
""
],
[
"Serrano",
"S.",
""
],
[
"Sirignano",
"C.",
""
],
[
"Steinwagner",
"J.",
""
],
[
"Tereno",
"I.",
""
],
[
"Toledo-Moreo",
"R.",
""
],
[
"Torradeflot",
"F.",
""
],
[
"Tutusaus",
"I.",
""
],
[
"Valenziano",
"L.",
""
],
[
"Vassallo",
"T.",
""
],
[
"Wang",
"Y.",
""
],
[
"Weller",
"J.",
""
],
[
"Zerbi",
"F. M.",
""
],
[
"Zucca",
"E.",
""
],
[
"Burigana",
"C.",
""
],
[
"Scottez",
"V.",
""
]
] |
2506.02746 | Lin Xie | Lin Xie and Hanyi Li | Solving the Pod Repositioning Problem with Deep Reinforced Adaptive Large Neighborhood Search | 14 pages, 2 figures, conference | null | null | null | cs.RO cs.AI math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Pod Repositioning Problem (PRP) in Robotic Mobile Fulfillment Systems (RMFS) involves selecting optimal storage locations for pods returning from pick stations. This work presents an improved solution method that integrates Adaptive Large Neighborhood Search (ALNS) with Deep Reinforcement Learning (DRL). A DRL agent dynamically selects destroy and repair operators and adjusts key parameters such as destruction degree and acceptance thresholds during the search. Specialized heuristics for both operators are designed to reflect PRP-specific characteristics, including pod usage frequency and movement costs. Computational results show that this DRL-guided ALNS outperforms traditional approaches such as cheapest-place, fixed-place, binary integer programming, and static heuristics. The method demonstrates strong solution quality and illustrating the benefit of learning-driven control within combinatorial optimization for warehouse systems. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:07:41 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Xie",
"Lin",
""
],
[
"Li",
"Hanyi",
""
]
] |
2506.02747 | Gennaro Ciampa | Gennaro Ciampa, Tommaso Cortopassi, Gianluca Crippa, Raffaele D'Ambrosio, Stefano Spirito | A priori error estimates for the $\theta$-method for the flow of nonsmooth velocity fields | null | null | null | null | math.AP cs.NA math.NA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Velocity fields with low regularity (below the Lipschitz threshold) naturally arise in many models from mathematical physics, such as the inhomogeneous incompressible Navier-Stokes equations, and play a fundamental role in the analysis of nonlinear PDEs. The DiPerna-Lions theory ensures existence and uniqueness of the flow associated with a divergence-free velocity field with Sobolev regularity. In this paper, we establish a priori error estimates showing a logarithmic rate of convergence of numerical solutions, constructed via the $\theta$-method, towards the exact (analytic) flow for a velocity field with Sobolev regularity. In addition, we derive analogous a priori error estimates for Lagrangian solutions of the associated transport equation, exhibiting the same logarithmic rate of convergence. Our theoretical results are supported by numerical experiments, which confirm the predicted logarithmic behavior. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:11:01 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Ciampa",
"Gennaro",
""
],
[
"Cortopassi",
"Tommaso",
""
],
[
"Crippa",
"Gianluca",
""
],
[
"D'Ambrosio",
"Raffaele",
""
],
[
"Spirito",
"Stefano",
""
]
] |
2506.02748 | Aditya Mahadeva Arabhavi | A. M. Arabhavi, I. Kamp, Th. Henning, E. F. van Dishoeck, H. Jang, L. B. F. M. Waters, V. Christiaens, D. Gasman, I. Pascucci, G. Perotti, S. L. Grant, M. G\"udel, P.-O. Lagage, D. Barrado, A. Caratti o Garatti, F. Lahuis, T. Kaeufer, J. Kanwar, M. Morales-Calder\'on, K. Schwarz, A. D. Sellek, B. Tabone, M. Temmink, M. Vlasblom, P. Patapis | MINDS: The very low-mass star and brown dwarf sample. Detections and trends in the inner disk gas | 20 pages, 18 figures, Accepted for publication in Astronomy & Astrophysics | null | null | null | astro-ph.EP astro-ph.GA astro-ph.SR | http://creativecommons.org/licenses/by/4.0/ | Planet-forming disks around brown dwarfs and very low-mass stars (VLMS) are on average less massive and are expected to undergo faster radial solid transport than their higher mass counterparts. Spitzer had detected C$_2$H$_2$, CO$_2$ and HCN around these objects. With better sensitivity and spectral resolving power, JWST recently revealed incredibly carbon-rich spectra from such disks. A study of a larger sample of objects is necessary to understand how common such carbon-rich inner disk regions are and to put constraints on their evolution. We present and analyze MIRI observations of 10 disks around VLMS from the MIRI GTO program. This sample is diverse, with the central object ranging in mass from 0.02 to 0.14 $M_{\odot}$. They are located in three star-forming regions and a moving group (1-10 Myr). We identify molecular emission in all sources and report detection rates. We compare the molecular flux ratios between different species and to dust emission strengths. We also compare the flux ratios with the stellar and disk properties. The spectra of these VLMS disks are extremely molecular rich, and we detect the 10 $\mu$m silicate dust emission feature in 70% of the sample. We detect C$_2$H$_2$ and HCN in all of the sources and find larger hydrocarbons such as C$_4$H$_2$ and C$_6$H$_6$ in nearly all sources. Among O-bearing molecules, we find firm detections of CO$_2$, H$_2$O, and CO in 90%, 50%, and 20% of the sample, respectively. We find that the detection rates of organic molecules correlate with other organic molecules and anti-correlate with inorganic molecules. Hydrocarbon-rich sources show a weaker 10$\mu$m dust strength as well as lower disk dust mass than the oxygen-rich sources. We find potential evidence for C/O enhancement with disk age. The observed trends are consistent with models that suggest rapid inward solid material transport and grain growth. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:11:27 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Arabhavi",
"A. M.",
""
],
[
"Kamp",
"I.",
""
],
[
"Henning",
"Th.",
""
],
[
"van Dishoeck",
"E. F.",
""
],
[
"Jang",
"H.",
""
],
[
"Waters",
"L. B. F. M.",
""
],
[
"Christiaens",
"V.",
""
],
[
"Gasman",
"D.",
""
],
[
"Pascucci",
"I.",
""
],
[
"Perotti",
"G.",
""
],
[
"Grant",
"S. L.",
""
],
[
"Güdel",
"M.",
""
],
[
"Lagage",
"P. -O.",
""
],
[
"Barrado",
"D.",
""
],
[
"Garatti",
"A. Caratti o",
""
],
[
"Lahuis",
"F.",
""
],
[
"Kaeufer",
"T.",
""
],
[
"Kanwar",
"J.",
""
],
[
"Morales-Calderón",
"M.",
""
],
[
"Schwarz",
"K.",
""
],
[
"Sellek",
"A. D.",
""
],
[
"Tabone",
"B.",
""
],
[
"Temmink",
"M.",
""
],
[
"Vlasblom",
"M.",
""
],
[
"Patapis",
"P.",
""
]
] |
2506.02749 | Changyi Xiao | Changyi Xiao, Yixin Cao | Knowledge Graph Completion by Intermediate Variables Regularization | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Knowledge graph completion (KGC) can be framed as a 3-order binary tensor completion task. Tensor decomposition-based (TDB) models have demonstrated strong performance in KGC. In this paper, we provide a summary of existing TDB models and derive a general form for them, serving as a foundation for further exploration of TDB models. Despite the expressiveness of TDB models, they are prone to overfitting. Existing regularization methods merely minimize the norms of embeddings to regularize the model, leading to suboptimal performance. Therefore, we propose a novel regularization method for TDB models that addresses this limitation. The regularization is applicable to most TDB models and ensures tractable computation. Our method minimizes the norms of intermediate variables involved in the different ways of computing the predicted tensor. To support our regularization method, we provide a theoretical analysis that proves its effect in promoting low trace norm of the predicted tensor to reduce overfitting. Finally, we conduct experiments to verify the effectiveness of our regularization technique as well as the reliability of our theoretical analysis. The code is available at https://github.com/changyi7231/IVR. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:11:33 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Xiao",
"Changyi",
""
],
[
"Cao",
"Yixin",
""
]
] |
2506.02750 | Yankai Chen | Yankai Chen and Yue Que and Xinni Zhang and Chen Ma and Irwin King | Learning Binarized Representations with Pseudo-positive Sample Enhancement for Efficient Graph Collaborative Filtering | Accepted by TOIS | null | null | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Learning vectorized embeddings is fundamental to many recommender systems for user-item matching. To enable efficient online inference, representation binarization, which embeds latent features into compact binary sequences, has recently shown significant promise in optimizing both memory usage and computational overhead. However, existing approaches primarily focus on numerical quantization, neglecting the associated information loss, which often results in noticeable performance degradation. To address these issues, we study the problem of graph representation binarization for efficient collaborative filtering. Our findings indicate that explicitly mitigating information loss at various stages of embedding binarization has a significant positive impact on performance. Building on these insights, we propose an enhanced framework, BiGeaR++, which specifically leverages supervisory signals from pseudo-positive samples, incorporating both real item data and latent embedding samples. Compared to its predecessor BiGeaR, BiGeaR++ introduces a fine-grained inference distillation mechanism and an effective embedding sample synthesis approach. Empirical evaluations across five real-world datasets demonstrate that the new designs in BiGeaR++ work seamlessly well with other modules, delivering substantial improvements of around 1%-10% over BiGeaR and thus achieving state-of-the-art performance compared to the competing methods. Our implementation is available at https://github.com/QueYork/BiGeaR-SS. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:11:43 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Chen",
"Yankai",
""
],
[
"Que",
"Yue",
""
],
[
"Zhang",
"Xinni",
""
],
[
"Ma",
"Chen",
""
],
[
"King",
"Irwin",
""
]
] |
2506.02751 | Chuanyu Fu | Chuanyu Fu, Yuqi Zhang, Kunbin Yao, Guanying Chen, Yuan Xiong, Chuan Huang, Shuguang Cui, Xiaochun Cao | RobustSplat: Decoupling Densification and Dynamics for Transient-Free 3DGS | Project page: https://fcyycf.github.io/RobustSplat/ | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | 3D Gaussian Splatting (3DGS) has gained significant attention for its real-time, photo-realistic rendering in novel-view synthesis and 3D modeling. However, existing methods struggle with accurately modeling scenes affected by transient objects, leading to artifacts in the rendered images. We identify that the Gaussian densification process, while enhancing scene detail capture, unintentionally contributes to these artifacts by growing additional Gaussians that model transient disturbances. To address this, we propose RobustSplat, a robust solution based on two critical designs. First, we introduce a delayed Gaussian growth strategy that prioritizes optimizing static scene structure before allowing Gaussian splitting/cloning, mitigating overfitting to transient objects in early optimization. Second, we design a scale-cascaded mask bootstrapping approach that first leverages lower-resolution feature similarity supervision for reliable initial transient mask estimation, taking advantage of its stronger semantic consistency and robustness to noise, and then progresses to high-resolution supervision to achieve more precise mask prediction. Extensive experiments on multiple challenging datasets show that our method outperforms existing methods, clearly demonstrating the robustness and effectiveness of our method. Our project page is https://fcyycf.github.io/RobustSplat/. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:13:48 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Fu",
"Chuanyu",
""
],
[
"Zhang",
"Yuqi",
""
],
[
"Yao",
"Kunbin",
""
],
[
"Chen",
"Guanying",
""
],
[
"Xiong",
"Yuan",
""
],
[
"Huang",
"Chuan",
""
],
[
"Cui",
"Shuguang",
""
],
[
"Cao",
"Xiaochun",
""
]
] |
2506.02752 | Hongpei Li | Hongpei Li, Ziyan He, Yufei Wang, Wenting Tu, Shanwen Pu, Qi Deng and Dongdong Ge | BenLOC: A Benchmark for Learning to Configure MIP Optimizers | A Benchmark for learning to configurate MIP Optimizers (Solvers) | null | null | null | math.OC | http://creativecommons.org/licenses/by/4.0/ | The automatic configuration of Mixed-Integer Programming (MIP) optimizers has become increasingly critical as the large number of configurations can significantly affect solver performance. Yet the lack of standardized evaluation frameworks has led to data leakage and over-optimistic claims, as prior studies often rely on homogeneous datasets and inconsistent experimental setups. To promote a fair evaluation process, we present BenLOC, a comprehensive benchmark and open-source toolkit, which not only offers an end-to-end pipeline for learning instance-wise MIP optimizer configurations, but also standardizes dataset selection, train-test splits, feature engineering and baseline choice for unbiased and comprehensive evaluations. Leveraging this framework, we conduct an empirical analysis on five well-established MIP datasets and compare classical machine learning models with handcrafted features against state-of-the-art deep-learning techniques. The results demonstrate the importance of datasets, features and baseline criteria proposed by BenLOC and the effectiveness of BenLOC in providing unbiased and comprehensive evaluations. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:16:24 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Li",
"Hongpei",
""
],
[
"He",
"Ziyan",
""
],
[
"Wang",
"Yufei",
""
],
[
"Tu",
"Wenting",
""
],
[
"Pu",
"Shanwen",
""
],
[
"Deng",
"Qi",
""
],
[
"Ge",
"Dongdong",
""
]
] |
2506.02753 | Aisha Alansari | Aisha Alansari, Hamzah Luqman | Multi-task Learning with Active Learning for Arabic Offensive Speech Detection | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The rapid growth of social media has amplified the spread of offensive, violent, and vulgar speech, which poses serious societal and cybersecurity concerns. Detecting such content in Arabic text is particularly complex due to limited labeled data, dialectal variations, and the language's inherent complexity. This paper proposes a novel framework that integrates multi-task learning (MTL) with active learning to enhance offensive speech detection in Arabic social media text. By jointly training on two auxiliary tasks, violent and vulgar speech, the model leverages shared representations to improve the detection accuracy of the offensive speech. Our approach dynamically adjusts task weights during training to balance the contribution of each task and optimize performance. To address the scarcity of labeled data, we employ an active learning strategy through several uncertainty sampling techniques to iteratively select the most informative samples for model training. We also introduce weighted emoji handling to better capture semantic cues. Experimental results on the OSACT2022 dataset show that the proposed framework achieves a state-of-the-art macro F1-score of 85.42%, outperforming existing methods while using significantly fewer fine-tuning samples. The findings of this study highlight the potential of integrating MTL with active learning for efficient and accurate offensive language detection in resource-constrained settings. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:17:03 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Alansari",
"Aisha",
""
],
[
"Luqman",
"Hamzah",
""
]
] |
2506.02754 | Luc Brogat-Motte | Luc Brogat-Motte, Alessandro Rudi, Riccardo Bonalli | Safely Learning Controlled Stochastic Dynamics | Under review at NeurIPS 2025 | null | null | null | stat.ML cs.LG | http://creativecommons.org/licenses/by/4.0/ | We address the problem of safely learning controlled stochastic dynamics from discrete-time trajectory observations, ensuring system trajectories remain within predefined safe regions during both training and deployment. Safety-critical constraints of this kind are crucial in applications such as autonomous robotics, finance, and biomedicine. We introduce a method that ensures safe exploration and efficient estimation of system dynamics by iteratively expanding an initial known safe control set using kernel-based confidence bounds. After training, the learned model enables predictions of the system's dynamics and permits safety verification of any given control. Our approach requires only mild smoothness assumptions and access to an initial safe control set, enabling broad applicability to complex real-world systems. We provide theoretical guarantees for safety and derive adaptive learning rates that improve with increasing Sobolev regularity of the true dynamics. Experimental evaluations demonstrate the practical effectiveness of our method in terms of safety, estimation accuracy, and computational efficiency. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:17:07 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Brogat-Motte",
"Luc",
""
],
[
"Rudi",
"Alessandro",
""
],
[
"Bonalli",
"Riccardo",
""
]
] |
2506.02755 | Soma Nishino | Soma Nishino | A central limit theorem for the stochastic cable equation | 26 pages | null | null | null | math.PR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we consider one-dimensional nonlinear stochastic cable equations driven by a multiplicative space-time white noise. Using the Malliavin--Stein's method, we prove the Gaussian fluctuation for the spatial average of the solution under the total variation distance with mild conditions. We also establish functional central limit theorem. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:21:34 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Nishino",
"Soma",
""
]
] |
2506.02756 | Ioanna Lykourentzou | Aaron Kessler, Tim Scheiber, Heinz Schmitz, Ioanna Lykourentzou | A Hierarchical Integer Linear Programming Approach for Optimizing Team Formation in Education | null | null | null | null | math.OC cs.CY | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Teamwork is integral to higher education, fostering students' interpersonal skills, improving learning outcomes, and preparing them for professional collaboration later in their careers. While team formation has traditionally been managed by humans, either instructors or students, algorithmic approaches have recently emerged to optimize this process. However, existing algorithmic team formation methods often focus on expert teams, overlook agency in choosing one's teammates, and are limited to a single team formation setting. These limitations make them less suitable for education, where no student can be left out, student agency is crucial for motivation, and team formation needs vary across courses and programs. In this paper, we introduce the EDUCATIONAL TEAM FORMATION problem (EDU-TF), a partitioning optimization problem model tailored to the unique needs of education, integrating both teacher and student requirements. To solve EDU-TF, we propose a modular optimization approach, one of the first to allow the flexible adjustment of objectives according to educational needs, enhancing the method's applicability across various classroom settings rather than just research environments. Results from evaluating ten strategies derived from our model on real-world university datasets indicate that our approach outperforms heuristic teacher-assigned teams by better accommodating student preferences. Our study contributes a new modular approach to partition-based algorithmic team formation and provides valuable insights for future research on team formation in educational settings. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:22:24 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Kessler",
"Aaron",
""
],
[
"Scheiber",
"Tim",
""
],
[
"Schmitz",
"Heinz",
""
],
[
"Lykourentzou",
"Ioanna",
""
]
] |
2506.02757 | Jinhan Liu | Ruiying Lu, Jinhan Liu, Chuan Du, Dandan Guo | Investigating Mask-aware Prototype Learning for Tabular Anomaly Detection | 12 pages, 11 figures | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Tabular anomaly detection, which aims at identifying deviant samples, has been crucial in a variety of real-world applications, such as medical disease identification, financial fraud detection, intrusion monitoring, etc. Although recent deep learning-based methods have achieved competitive performances, these methods suffer from representation entanglement and the lack of global correlation modeling, which hinders anomaly detection performance. To tackle the problem, we incorporate mask modeling and prototype learning into tabular anomaly detection. The core idea is to design learnable masks by disentangled representation learning within a projection space and extracting normal dependencies as explicit global prototypes. Specifically, the overall model involves two parts: (i) During encoding, we perform mask modeling in both the data space and projection space with orthogonal basis vectors for learning shared disentangled normal patterns; (ii) During decoding, we decode multiple masked representations in parallel for reconstruction and learn association prototypes to extract normal characteristic correlations. Our proposal derives from a distribution-matching perspective, where both projection space learning and association prototype learning are formulated as optimal transport problems, and the calibration distances are utilized to refine the anomaly scores. Quantitative and qualitative experiments on 20 tabular benchmarks demonstrate the effectiveness and interpretability of our model. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:22:44 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Lu",
"Ruiying",
""
],
[
"Liu",
"Jinhan",
""
],
[
"Du",
"Chuan",
""
],
[
"Guo",
"Dandan",
""
]
] |
2506.02758 | Stefano Bann\`o | Stefano Bann\`o, Kate Knill, Mark Gales | Exploiting the English Vocabulary Profile for L2 word-level vocabulary assessment with LLMs | Accepted to the 20th Workshop on Innovative Use of NLP for Building Educational Applications | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Vocabulary use is a fundamental aspect of second language (L2) proficiency. To date, its assessment by automated systems has typically examined the context-independent, or part-of-speech (PoS) related use of words. This paper introduces a novel approach to enable fine-grained vocabulary evaluation exploiting the precise use of words within a sentence. The scheme combines large language models (LLMs) with the English Vocabulary Profile (EVP). The EVP is a standard lexical resource that enables in-context vocabulary use to be linked with proficiency level. We evaluate the ability of LLMs to assign proficiency levels to individual words as they appear in L2 learner writing, addressing key challenges such as polysemy, contextual variation, and multi-word expressions. We compare LLMs to a PoS-based baseline. LLMs appear to exploit additional semantic information that yields improved performance. We also explore correlations between word-level proficiency and essay-level proficiency. Finally, the approach is applied to examine the consistency of the EVP proficiency levels. Results show that LLMs are well-suited for the task of vocabulary assessment. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:23:57 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Bannò",
"Stefano",
""
],
[
"Knill",
"Kate",
""
],
[
"Gales",
"Mark",
""
]
] |
2506.02759 | Evan Cavallo | Evan Cavallo, Christian Sattler | The algebraic small object argument as a saturation | null | null | null | null | math.CT math.AT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We analyze the structure of left maps in algebraic weak factorization systems constructed using Garner's algebraic small object argument. We find that any left map can be constructed from generators in Bourke and Garner's double category of left maps by operations that parallel the classical cell-complex-forming operations of Quillen's small object argument (coproducts, cobase changes, transfinite composites, and retracts). Our main theorems are phrased as "saturation" principles, which express the closure conditions necessary for a given property or structure to extend from generators to all left maps. The core of the argument is an analysis of the construction of the free monad on a pointed endofunctor. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:24:18 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Cavallo",
"Evan",
""
],
[
"Sattler",
"Christian",
""
]
] |
2506.02760 | Hakime BarghiZanjani | Hakime BarghiZanjani, Bikshapathi Gouda, and Antti T\"olli | Coordinated Multi-BS SSB Beam Design for Enhanced Initial Access Coverage | This paper has been submitted to IEEE GLOBECOM 2025. It is a 6-page submission prepared according to the conference guidelines | null | null | null | eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Ensuring strong synchronization signal block (SSB) coverage is essential for reliable user equipment (UE) connection during initial access. While techniques such as power boosting and network densification are commonly used, this work explores joint transmission (JT) of SSBs as an alternative to enhance coverage. Although JT is widely applied in data transmission, its use for SSBs has not been explored due to the lack of channel state information, which prevents coherent signal alignment across base stations (BSs). To address this, we propose a repetition-based JT strategy using a small set of predefined phase configurations at the BSs. This enables the UE to coherently combine multiple SSB receptions and achieve constructive gain regardless of its location. To reduce overhead, a limited number of joint beam configurations is selected to maximize the coverage. Simulation results under the line-of-sight conditions show up to 6 dB relative SNR gain with JT of SSBs using 4 BSs, compared to independent SSB transmission under the same resource budget. These results highlight the potential of JT to improve the coverage of SSBs during the initial access. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:24:22 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"BarghiZanjani",
"Hakime",
""
],
[
"Gouda",
"Bikshapathi",
""
],
[
"Tölli",
"Antti",
""
]
] |
2506.02761 | Renyang Liu | Renyang Liu, Wenjie Feng, Tianwei Zhang, Wei Zhou, Xueqi Cheng, See-Kiong Ng | Rethinking Machine Unlearning in Image Generation Models | Accepted by ACM CCS 2025 | ACM Conference on Computer and Communications Security (CCS 2025) | null | null | cs.AI cs.CL cs.CR cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | With the surge and widespread application of image generation models, data privacy and content safety have become major concerns and attracted great attention from users, service providers, and policymakers. Machine unlearning (MU) is recognized as a cost-effective and promising means to address these challenges. Despite some advancements, image generation model unlearning (IGMU) still faces remarkable gaps in practice, e.g., unclear task discrimination and unlearning guidelines, lack of an effective evaluation framework, and unreliable evaluation metrics. These can hinder the understanding of unlearning mechanisms and the design of practical unlearning algorithms. We perform exhaustive assessments over existing state-of-the-art unlearning algorithms and evaluation standards, and discover several critical flaws and challenges in IGMU tasks. Driven by these limitations, we make several core contributions, to facilitate the comprehensive understanding, standardized categorization, and reliable evaluation of IGMU. Specifically, (1) We design CatIGMU, a novel hierarchical task categorization framework. It provides detailed implementation guidance for IGMU, assisting in the design of unlearning algorithms and the construction of testbeds. (2) We introduce EvalIGMU, a comprehensive evaluation framework. It includes reliable quantitative metrics across five critical aspects. (3) We construct DataIGM, a high-quality unlearning dataset, which can be used for extensive evaluations of IGMU, training content detectors for judgment, and benchmarking the state-of-the-art unlearning algorithms. With EvalIGMU and DataIGM, we discover that most existing IGMU algorithms cannot handle the unlearning well across different evaluation dimensions, especially for preservation and robustness. Code and models are available at https://github.com/ryliu68/IGMU. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:25:14 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Liu",
"Renyang",
""
],
[
"Feng",
"Wenjie",
""
],
[
"Zhang",
"Tianwei",
""
],
[
"Zhou",
"Wei",
""
],
[
"Cheng",
"Xueqi",
""
],
[
"Ng",
"See-Kiong",
""
]
] |
2506.02762 | Ryo Tatsumi | Ryo Tatsumi, Takahiro Chiba, Takashi Komine, Hiroaki Matsueda | Chaotic magnetization dynamics in magnetic Duffing oscillator | 12 pages, 10 figures | Phys. Rev. E, 111, 064202 (2025) | 10.1103/PhysRevE.111.064202 | null | cond-mat.mes-hall cond-mat.mtrl-sci nlin.CD physics.app-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We propose a magnetic analogy of the Duffing oscillator--magnetic Duffing oscillator--which is characterized by a double-well magnetic potential of a ferromagnet with a uniaxial magnetic anisotropy. Based on the linear stability analysis of the Landau-Lifshitz-Gilbert equation, we show that an external magnetic field applied perpendicular to the magnetic anisotropy field creates an anharmonicity on the magnetic potential, generating homoclinic orbits in the phase space. By evaluating the Lyapunov exponent, we demonstrate that the magnetic Duffing oscillator exhibits chaotic behaviors in the presence of periodically oscillating external forces: Oersted field and spin-orbit torque by considering the ferromagnet/heavy-metal bilayer. We also show that the external magnetic field can be adjusted to generate or modify homoclinic orbits, thereby controlling the parameter range of the oscillating external forces that induce chaos. This work deepens our understanding of chaotic magnetization dynamics by bridging the fields of nonlinear dynamics and spintronics. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:26:33 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Tatsumi",
"Ryo",
""
],
[
"Chiba",
"Takahiro",
""
],
[
"Komine",
"Takashi",
""
],
[
"Matsueda",
"Hiroaki",
""
]
] |
2506.02763 | Meng Zhang | Meng Zhang, Maosheng Xiang, Yuan-Sen Ting, Anish Maynur Amarsi, Hua-Wei Zhang, Jianrong Shi, Haibo Yuan, Haining Li, Jiahui Wang, Yaqian Wu, Tianmin Wu, Lanya Mou, Hong-liang Yan, Jifeng Liu | Homogeneous Stellar Atmospheric Parameters and 22 Elemental Abundances for FGK Stars Derived From LAMOST Low-resolution Spectra with DD-Payne | 36 pages, 28 figures, 4 tables. Accepted for publication in ApJS | null | 10.3847/1538-4365/add016 | null | astro-ph.SR astro-ph.GA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A deep understanding of our Galaxy desires detailed decomposition of its stellar populations via their chemical fingerprints. This requires precise stellar abundances of many elements for a large number of stars. Here we present an updated catalog of stellar labels derived from LAMOST low-resolution spectra in a physics-sensible and rigorous manner with DD-Payne, taking labels from high-resolution spectroscopy as training set. The catalog contains atmospheric parameters for 6.4 million stars released in LAMOST DR9, and abundances for 22 elements, namely, C, N, O, Na, Mg, Al, Si, Ca, Ti, Cr, Mn, Fe, Ni, Sr, Y, Zr, Ba, La, Ce, Nd, Sm, and Eu, for nearly 3.6 million stars with spectral signal-to-noise ratio (SNR) higher than 20. The [Fe/H] is valid down to $\sim$-4.0, while elemental abundance ratios [X/Fe] are mostly valid for stars with [Fe/H] $\gtrsim-2.0$. Measurement errors in these labels are sensitive to and almost inversely proportional with SNR. For stars with S/N>50, we achieved a typical error of 30 K in Teff, 0.07 dex in $\log g$, $\sim0.05$ dex in abundances for most elements with atomic number smaller than Sr, and 0.1--0.2 dex for heavier elements. Homogenization to the label estimates is carried out via dedicated internal and external calibration. In particular, the non-local thermal equilibrium effect is corrected for the [Fe/H] estimates, the Teff is calibrated to the infrared flux method scale, and the $\log~g$ is validated with asteroseismic measurements. The elemental abundances are internally calibrated using wide binaries, eliminating systematic trend with effective temperature. The catalog is publicly available. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:28:06 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Zhang",
"Meng",
""
],
[
"Xiang",
"Maosheng",
""
],
[
"Ting",
"Yuan-Sen",
""
],
[
"Amarsi",
"Anish Maynur",
""
],
[
"Zhang",
"Hua-Wei",
""
],
[
"Shi",
"Jianrong",
""
],
[
"Yuan",
"Haibo",
""
],
[
"Li",
"Haining",
""
],
[
"Wang",
"Jiahui",
""
],
[
"Wu",
"Yaqian",
""
],
[
"Wu",
"Tianmin",
""
],
[
"Mou",
"Lanya",
""
],
[
"Yan",
"Hong-liang",
""
],
[
"Liu",
"Jifeng",
""
]
] |
2506.02764 | Fatma Youssef Mohammed | Fatma Youssef Mohammed and Kostas Alexis | Unified Attention Modeling for Efficient Free-Viewing and Visual Search via Shared Representations | Accepted to the 2025 IEEE International Conference on Development and Learning (ICDL) | null | null | null | cs.CV cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Computational human attention modeling in free-viewing and task-specific settings is often studied separately, with limited exploration of whether a common representation exists between them. This work investigates this question and proposes a neural network architecture that builds upon the Human Attention transformer (HAT) to test the hypothesis. Our results demonstrate that free-viewing and visual search can efficiently share a common representation, allowing a model trained in free-viewing attention to transfer its knowledge to task-driven visual search with a performance drop of only 3.86% in the predicted fixation scanpaths, measured by the semantic sequence score (SemSS) metric which reflects the similarity between predicted and human scanpaths. This transfer reduces computational costs by 92.29% in terms of GFLOPs and 31.23% in terms of trainable parameters. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:29:11 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Mohammed",
"Fatma Youssef",
""
],
[
"Alexis",
"Kostas",
""
]
] |
2506.02765 | Chunwei Tian | Chunwei Tian, Kai Liu, Bob Zhang, Zhixiang Huang, Chia-Wen Lin and David Zhang | A Dynamic Transformer Network for Vehicle Detection | 8 pages, 5 figures. This paper has been accepted for publication in IEEE Transactions on Consumer Electronics | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Stable consumer electronic systems can assist traffic better. Good traffic consumer electronic systems require collaborative work between traffic algorithms and hardware. However, performance of popular traffic algorithms containing vehicle detection methods based on deep networks via learning data relation rather than learning differences in different lighting and occlusions is limited. In this paper, we present a dynamic Transformer network for vehicle detection (DTNet). DTNet utilizes a dynamic convolution to guide a deep network to dynamically generate weights to enhance adaptability of an obtained detector. Taking into relations of different information account, a mixed attention mechanism based channel attention and Transformer is exploited to strengthen relations of channels and pixels to extract more salient information for vehicle detection. To overcome the drawback of difference in an image account, a translation-variant convolution relies on spatial location information to refine obtained structural information for vehicle detection. Experimental results illustrate that our DTNet is competitive for vehicle detection. Code of the proposed DTNet can be obtained at https://github.com/hellloxiaotian/DTNet. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:29:35 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Tian",
"Chunwei",
""
],
[
"Liu",
"Kai",
""
],
[
"Zhang",
"Bob",
""
],
[
"Huang",
"Zhixiang",
""
],
[
"Lin",
"Chia-Wen",
""
],
[
"Zhang",
"David",
""
]
] |
2506.02766 | Agnese Seminara | Luca Gagliardi and Agnese Seminara | Learning to crawl: benefits and limits of centralized vs distributed control | null | null | null | null | physics.bio-ph | http://creativecommons.org/licenses/by/4.0/ | We present a model of a crawler consisting of several suckers distributed along a straight line and connected by springs. Both proprioception and control are binary: the system responds to the elongation vs compression of its springs and the suckers can either adhere or remain idle. A central pattern generator delivers a traveling wave of compression, and the crawler is tasked to learn how to control adhesion to effectively ride the wave in order to crawl. We ask what are the benefits and limitations of distributed vs centralized learning architectures. Using tabular Q-learning we demonstrate that crawling can be learned by trial and error in a purely distributed setting where each sucker learns independently how to control adhesion, with no exchange of information among them. We find that centralizing control of all suckers enhances speed and robustness to failure by orchestrating smoother collective dynamics and partly overcoming the limitations of rudimentary proprioception and control. However, since the computational cost scales exponentially with the number of suckers, centralized control quickly becomes untreatable with the size of the crawler. We then show that intermediate levels of centralization, where multiple control centers coordinate subsets of suckers, can negotiate fast and robust crawling while avoiding excessive computational burden. Our model can be further generalized to explore the trade-offs between crawling speed, robustness to failure, computational cost and information exchange that shape the emergence of biological solutions for crawling and could inspire the design of robotic crawlers. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:30:14 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Gagliardi",
"Luca",
""
],
[
"Seminara",
"Agnese",
""
]
] |
2506.02767 | Marco Cal\`i | Marco Cal\`i, Giulio Giacomuzzo, Ruggero Carli and Alberto Dalla Libera | Accelerating Model-Based Reinforcement Learning using Non-Linear Trajectory Optimization | null | null | null | null | cs.LG cs.RO | http://creativecommons.org/licenses/by/4.0/ | This paper addresses the slow policy optimization convergence of Monte Carlo Probabilistic Inference for Learning Control (MC-PILCO), a state-of-the-art model-based reinforcement learning (MBRL) algorithm, by integrating it with iterative Linear Quadratic Regulator (iLQR), a fast trajectory optimization method suitable for nonlinear systems. The proposed method, Exploration-Boosted MC-PILCO (EB-MC-PILCO), leverages iLQR to generate informative, exploratory trajectories and initialize the policy, significantly reducing the number of required optimization steps. Experiments on the cart-pole task demonstrate that EB-MC-PILCO accelerates convergence compared to standard MC-PILCO, achieving up to $\bm{45.9\%}$ reduction in execution time when both methods solve the task in four trials. EB-MC-PILCO also maintains a $\bm{100\%}$ success rate across trials while solving the task faster, even in cases where MC-PILCO converges in fewer iterations. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:30:59 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Calì",
"Marco",
""
],
[
"Giacomuzzo",
"Giulio",
""
],
[
"Carli",
"Ruggero",
""
],
[
"Libera",
"Alberto Dalla",
""
]
] |
2506.02768 | Eytan Canzini | Ethan Canzini, Simon Pope, Ashutosh Tiwari | Geometric Visual Servo Via Optimal Transport | 19 pages, 5 figures | null | null | null | cs.RO cs.SY eess.SY | http://creativecommons.org/licenses/by/4.0/ | When developing control laws for robotic systems, the principle factor when examining their performance is choosing inputs that allow smooth tracking to a reference input. In the context of robotic manipulation, this involves translating an object or end-effector from an initial pose to a target pose. Robotic manipulation control laws frequently use vision systems as an error generator to track features and produce control inputs. However, current control algorithms don't take into account the probabilistic features that are extracted and instead rely on hand-tuned feature extraction methods. Furthermore, the target features can exist in a static pose thus allowing a combined pose and feature error for control generation. We present a geometric control law for the visual servoing problem for robotic manipulators. The input from the camera constitutes a probability measure on the 3-dimensional Special Euclidean task-space group, where the Wasserstein distance between the current and desired poses is analogous with the geometric geodesic. From this, we develop a controller that allows for both pose and image-based visual servoing by combining classical PD control with gravity compensation with error minimization through the use of geodesic flows on a 3-dimensional Special Euclidean group. We present our results on a set of test cases demonstrating the generalisation ability of our approach to a variety of initial positions. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:38:09 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Canzini",
"Ethan",
""
],
[
"Pope",
"Simon",
""
],
[
"Tiwari",
"Ashutosh",
""
]
] |
2506.02769 | Andrei Manolescu | Anna Sitek, Tudor Gabriel Dumitru, Sigurdur I. Erlingsson, Andrei Manolescu | Spin-orbit interaction in square core-shell nanowires | 25th International Conference on Transparent Optical Networks (ICTON), Barcelona, July 2025 | null | null | null | cond-mat.mes-hall | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We theoretically investigate the spin-orbit interaction of electrons confined in the outer regions of square core-shell nanowires. The polygonal cross section leads to the accumulation of low-energy electrons in the corners and the formation of a significant energy gap that separates these corner-localized states from higher-energy states localized along the sides. We show that the low-energy states behave like the states of independent quantum wires, while the higher-energy states exhibit features characteristic of coupled wires. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:41:05 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Sitek",
"Anna",
""
],
[
"Dumitru",
"Tudor Gabriel",
""
],
[
"Erlingsson",
"Sigurdur I.",
""
],
[
"Manolescu",
"Andrei",
""
]
] |
2506.02770 | Pierrick Bousseau | H\"ulya Arg\"uz and Pierrick Bousseau | BPS polynomials and Welschinger invariants | 48 pages, 10 figures. Comments welcome! | null | null | null | math.AG hep-th math.SG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We generalize Block-G\"ottsche polynomials, originally defined for toric del Pezzo surfaces, to arbitrary surfaces. To do this, we show that these polynomials arise as special cases of BPS polynomials, defined for any surface $S$ as Laurent polynomials in a formal variable $q$ encoding the BPS invariants of the $3$-fold $S \times \mathbb{P}^1$. We conjecture that for surfaces $S_n$ obtained by blowing up $\mathbb{P}^2$ at $n$ general points, the evaluation of BPS polynomials at $q=-1$ yields Welschinger invariants, given by signed counts of real rational curves. We prove this conjecture for all surfaces $S_n$ with $n \leq 6$. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:41:38 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Argüz",
"Hülya",
""
],
[
"Bousseau",
"Pierrick",
""
]
] |
2506.02771 | Taufik Abrao PhD | Bruno Felipe Costa and Taufik Abr\~ao | Derivation of CRB and Refined SINR Expressions for OTFS-RSMA LEO ISAC Systems | 5 pages, 28 equations, 0 figures | null | null | null | eess.SP | http://creativecommons.org/licenses/by/4.0/ | This document provides detailed step-by-step derivations for the Cram\'er-Rao Bounds (CRB) for sensing parameters and the refined Signal-to-Interference-plus-Noise Ratio (SINR) expressions under imperfect Channel State Information (CSI) and imperfect Successive Interference Cancellation (SIC) for the Orthogonal Time Frequency Space (OTFS) Rate-Splitting Multiple Access (RSMA) framework presented in our main work "An Integrated OTFS-RSMA Framework for LEO Satellite ISAC: Modeling, Metrics, and Potential". These derivations support the analytical expressions and models in the broad main discussion. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:42:00 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Costa",
"Bruno Felipe",
""
],
[
"Abrão",
"Taufik",
""
]
] |
2506.02772 | Ronald Mahler | Ronald Mahler | Generalized Labeled Multi-Bernoulli Filters and Multitarget-Correlation Models | null | null | null | null | stat.ME stat.CO | http://creativecommons.org/licenses/by/4.0/ | The generalized labeled multi-Bernoulli (GLMB) filter is a theoretically rigorous Bayes-optimal multitarget tracking algorithm with computationally tractable implementations, based on labeled random finite set (LRFS) theory. It presumes that multitarget populations can be approximated using GLMB multitarget probability density functions (p.d.f.'s), which consist of weighted hypotheses regarding the current target-states. A special case of the GLMB p.d.f.-the LMB p.d.f.-presumes that the targets are statistically independent. This paper demonstrates that a) GLMB p.d.f.'s can be interpreted as straightforward generalizations of LMB p.d.f.'s to statistically correlated target populations, given an implicit presumption of "simple labeled correlation" (SLC) models of multitarget correlation; b) the GLMB filter can be reformulated as a SLC-GLMB filter; and c) SLC models seem primarily appropriate for target clusters consisting of small numbers of closely-spaced targets. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:46:47 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Mahler",
"Ronald",
""
]
] |
2506.02773 | He Kong Dr. | Linya Fu, Yu Liu, Zhijie Liu, Zedong Yang, Zhong-Qiu Wang, Youfu Li, and He Kong | AuralNet: Hierarchical Attention-based 3D Binaural Localization of Overlapping Speakers | Accepted and to appear at Interspeech 2025 | null | null | null | eess.AS cs.SD | http://creativecommons.org/licenses/by/4.0/ | We propose AuralNet, a novel 3D multi-source binaural sound source localization approach that localizes overlapping sources in both azimuth and elevation without prior knowledge of the number of sources. AuralNet employs a gated coarse-tofine architecture, combining a coarse classification stage with a fine-grained regression stage, allowing for flexible spatial resolution through sector partitioning. The model incorporates a multi-head self-attention mechanism to capture spatial cues in binaural signals, enhancing robustness in noisy-reverberant environments. A masked multi-task loss function is designed to jointly optimize sound detection, azimuth, and elevation estimation. Extensive experiments in noisy-reverberant conditions demonstrate the superiority of AuralNet over recent methods | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:48:28 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Fu",
"Linya",
""
],
[
"Liu",
"Yu",
""
],
[
"Liu",
"Zhijie",
""
],
[
"Yang",
"Zedong",
""
],
[
"Wang",
"Zhong-Qiu",
""
],
[
"Li",
"Youfu",
""
],
[
"Kong",
"He",
""
]
] |
2506.02775 | Bogna Kubik | B. Kubik, R. Barbier, G. Smadja, S. Ferriol, Y. Conseil, Y. Copin, W. Gillard, S. Dusini, K. Jahnke, E. Prieto, N. Auricchio, E. Balbi, A. Balestra, P. Battaglia, V. Capobianco, R. Chary, L. Corcione, F. Cogato, G. Delucchi, E. Franceschi, L. Gabarra, F. Gianotti, F. Grupp, E. Lentini, S. Ligori, E. Medinaceli, G. Morgante, K. Paterson, E. Romelli, L. Sauniere, M. Schirmer, C. Sirignano G. Testera, M. Trifoglio, A. Troja, L. Valenziano, M. Frailis, M. Scodeggio, J.-C. Barriere, M. Berthe, C. Bodendorf, A. Caillat, M. Carle, R. Casas, H. Cho, A. Costille, F. Ducret, B. Garilli, W. Holmes, F. Hormuth, A. Hornstrup, M. Jhabvala, R. Kohley, D. Le Mignant, P. B. Lilje, I. Lloro, C. Padilla, G. Polenta, J.-C. Salvignol, G. Seidel, B. Serra, A. Secroun, L. Stanco, R. Toledo-Moreo, S. Anselmi, E. Borsato, L. Caillat, C. Colodro-Conde, V. Conforti, J. E. Davies, A. Renzi, F. Dal Corso, S. Davini, A. Derosa, J. J. Diaz, S. Di Domizio, D. Di Ferdinando, R. Farinelli, A. G. Ferrari, F. Fornari, F. Giacomini, O. Krause, F. Laudisio, J. Macias-Perez, J. Marpaud, N. Mauri, R. da Silva, M. Niclas, F. Passalacqua, I. Risso, P. Lagier, A. N. Sorensen, P. Stassi, J. Steinwagner, M. Tenti, C. Thizy, S. Tosi, R. Travaglini, O. Tubio, C. Valieri, S. Ventura, C. Vescovi, and J. Zoubian | Performance of the image persistence model for Euclid infrared detectors | SPIE Astronomical Telescopes + Instrumentation, 2024, Yokohama, Japan | Proceedings Volume 13103, X-Ray, Optical, and Infrared Detectors for Astronomy XI; 1310315 (2024) | 10.1117/12.3014312 | null | astro-ph.IM | http://creativecommons.org/licenses/by/4.0/ | Large-format infrared detectors are at the heart of major ground and space-based astronomical instruments, and the HgCdTe HxRG is the most widely used. The Near Infrared Spectrometer and Photometer (NISP) of the ESA's Euclid mission launched in July 2023 hosts 16 H2RG detectors in the focal plane. Their performance relies heavily on the effect of image persistence, which results in residual images that can remain in the detector for a long time contaminating any subsequent observations. Deriving a precise model of image persistence is challenging due to the sensitivity of this effect to observation history going back hours or even days. Nevertheless, persistence removal is a critical part of image processing because it limits the accuracy of the derived cosmological parameters. We will present the empirical model of image persistence derived from ground characterization data, adapted to the Euclid observation sequence and compared with the data obtained during the in-orbit calibrations of the satellite. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:51:49 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Kubik",
"B.",
""
],
[
"Barbier",
"R.",
""
],
[
"Smadja",
"G.",
""
],
[
"Ferriol",
"S.",
""
],
[
"Conseil",
"Y.",
""
],
[
"Copin",
"Y.",
""
],
[
"Gillard",
"W.",
""
],
[
"Dusini",
"S.",
""
],
[
"Jahnke",
"K.",
""
],
[
"Prieto",
"E.",
""
],
[
"Auricchio",
"N.",
""
],
[
"Balbi",
"E.",
""
],
[
"Balestra",
"A.",
""
],
[
"Battaglia",
"P.",
""
],
[
"Capobianco",
"V.",
""
],
[
"Chary",
"R.",
""
],
[
"Corcione",
"L.",
""
],
[
"Cogato",
"F.",
""
],
[
"Delucchi",
"G.",
""
],
[
"Franceschi",
"E.",
""
],
[
"Gabarra",
"L.",
""
],
[
"Gianotti",
"F.",
""
],
[
"Grupp",
"F.",
""
],
[
"Lentini",
"E.",
""
],
[
"Ligori",
"S.",
""
],
[
"Medinaceli",
"E.",
""
],
[
"Morgante",
"G.",
""
],
[
"Paterson",
"K.",
""
],
[
"Romelli",
"E.",
""
],
[
"Sauniere",
"L.",
""
],
[
"Schirmer",
"M.",
""
],
[
"Testera",
"C. Sirignano G.",
""
],
[
"Trifoglio",
"M.",
""
],
[
"Troja",
"A.",
""
],
[
"Valenziano",
"L.",
""
],
[
"Frailis",
"M.",
""
],
[
"Scodeggio",
"M.",
""
],
[
"Barriere",
"J. -C.",
""
],
[
"Berthe",
"M.",
""
],
[
"Bodendorf",
"C.",
""
],
[
"Caillat",
"A.",
""
],
[
"Carle",
"M.",
""
],
[
"Casas",
"R.",
""
],
[
"Cho",
"H.",
""
],
[
"Costille",
"A.",
""
],
[
"Ducret",
"F.",
""
],
[
"Garilli",
"B.",
""
],
[
"Holmes",
"W.",
""
],
[
"Hormuth",
"F.",
""
],
[
"Hornstrup",
"A.",
""
],
[
"Jhabvala",
"M.",
""
],
[
"Kohley",
"R.",
""
],
[
"Mignant",
"D. Le",
""
],
[
"Lilje",
"P. B.",
""
],
[
"Lloro",
"I.",
""
],
[
"Padilla",
"C.",
""
],
[
"Polenta",
"G.",
""
],
[
"Salvignol",
"J. -C.",
""
],
[
"Seidel",
"G.",
""
],
[
"Serra",
"B.",
""
],
[
"Secroun",
"A.",
""
],
[
"Stanco",
"L.",
""
],
[
"Toledo-Moreo",
"R.",
""
],
[
"Anselmi",
"S.",
""
],
[
"Borsato",
"E.",
""
],
[
"Caillat",
"L.",
""
],
[
"Colodro-Conde",
"C.",
""
],
[
"Conforti",
"V.",
""
],
[
"Davies",
"J. E.",
""
],
[
"Renzi",
"A.",
""
],
[
"Corso",
"F. Dal",
""
],
[
"Davini",
"S.",
""
],
[
"Derosa",
"A.",
""
],
[
"Diaz",
"J. J.",
""
],
[
"Di Domizio",
"S.",
""
],
[
"Di Ferdinando",
"D.",
""
],
[
"Farinelli",
"R.",
""
],
[
"Ferrari",
"A. G.",
""
],
[
"Fornari",
"F.",
""
],
[
"Giacomini",
"F.",
""
],
[
"Krause",
"O.",
""
],
[
"Laudisio",
"F.",
""
],
[
"Macias-Perez",
"J.",
""
],
[
"Marpaud",
"J.",
""
],
[
"Mauri",
"N.",
""
],
[
"da Silva",
"R.",
""
],
[
"Niclas",
"M.",
""
],
[
"Passalacqua",
"F.",
""
],
[
"Risso",
"I.",
""
],
[
"Lagier",
"P.",
""
],
[
"Sorensen",
"A. N.",
""
],
[
"Stassi",
"P.",
""
],
[
"Steinwagner",
"J.",
""
],
[
"Tenti",
"M.",
""
],
[
"Thizy",
"C.",
""
],
[
"Tosi",
"S.",
""
],
[
"Travaglini",
"R.",
""
],
[
"Tubio",
"O.",
""
],
[
"Valieri",
"C.",
""
],
[
"Ventura",
"S.",
""
],
[
"Vescovi",
"C.",
""
],
[
"Zoubian",
"J.",
""
]
] |
2506.02776 | Emilio Mart\'inez-Pa\~neda | A. Lhoest, S. Kovacevic, D. Nguyen-Manh, J. Lim, E. Mart\'inez-Pa\~neda, M. Wenman | A mesoscale phase-field model of intergranular liquid lithium corrosion of ferritic/martensitic steels | null | null | null | null | physics.chem-ph cond-mat.mtrl-sci cs.CE physics.app-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A phase-field model is developed to simulate intergranular corrosion of ferritic/martensitic steels exposed to liquid lithium. The chromium concentration of the material is used to track the mass transport within the metal and liquid (corrosive) phase. The framework naturally captures intergranular corrosion by enhancing the diffusion of chromium along grain boundaries relative to the grain bulk with no special treatment for the corrosion front evolution. The formulation applies to arbitrary 2D and 3D polycrystalline geometries. The framework reproduces experimental measurements of weight loss and corrosion depth for a 9 wt\% Cr ferritic/martensitic steel exposed to static lithium at 600 $^\circ$C. A sensitivity analysis, varying near-surface grain density, grain size, and chromium depletion thickness, highlights the microstructural influence in the corrosion process. Moreover, the significance of saturation is considered and evaluated. Simulation results show that near-surface grain density is a deciding factor, whereas grain size dictates the susceptibility to intergranular corrosion. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:54:59 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Lhoest",
"A.",
""
],
[
"Kovacevic",
"S.",
""
],
[
"Nguyen-Manh",
"D.",
""
],
[
"Lim",
"J.",
""
],
[
"Martínez-Pañeda",
"E.",
""
],
[
"Wenman",
"M.",
""
]
] |
2506.02777 | Paul Maria Reuter | Paul M. Reuter, Michael Jessen | On the influence of language similarity in non-target speaker verification trials | accepted to Interspeech 2025 | null | null | null | eess.AS cs.SD | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we investigate the influence of language similarity in cross-lingual non-target speaker verification trials using a state-of-the-art speaker verification system, ECAPA-TDNN, trained on multilingual and monolingual variants of the VoxCeleb dataset. Our analysis of the score distribution patterns on multilingual Globalphone and LDC CTS reveals a clustering effect in speaker comparisons involving a training language, whereby the choice of comparison language only minimally impacts scores. Conversely, we observe a language similarity effect in trials involving languages not included in the training set of the speaker verification system, with scores correlating with language similarity measured by a language classification system, especially when using multilingual training data. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 11:56:52 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Reuter",
"Paul M.",
""
],
[
"Jessen",
"Michael",
""
]
] |
2506.02780 | Peiding Wang | Peiding Wang, Li Zhang, Fang Liu, Yinghao Zhu, Wang Xu, Lin Shi, Xiaoli Lian, Minxiao Li, Bo Shen, An Fu | Reuse or Generate? Accelerating Code Editing via Edit-Oriented Speculative Decoding | null | null | null | null | cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large Language Models (LLMs) have demonstrated remarkable capabilities in code editing, substantially enhancing software development productivity. However, the inherent complexity of code editing tasks forces existing approaches to rely on LLMs' autoregressive end-to-end generation, where decoding speed plays a critical role in efficiency. While inference acceleration techniques like speculative decoding are applied to improve the decoding efficiency, these methods fail to account for the unique characteristics of code editing tasks where changes are typically localized and existing code segments are reused. To address this limitation, we propose EfficientEdit, a novel method that improves LLM-based code editing efficiency through two key mechanisms based on speculative decoding: (1) effective reuse of original code segments while identifying potential edit locations, and (2) efficient generate edit content via high-quality drafts from edit-oriented draft models and a dynamic verification mechanism that balances quality and acceleration. Experimental results show that EfficientEdit can achieve up to 10.38$\times$ and 13.09$\times$ speedup compared to standard autoregressive decoding in CanItEdit and CodeIF-Bench, respectively, outperforming state-of-the-art inference acceleration approaches by up to 90.6%. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 12:01:20 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Wang",
"Peiding",
""
],
[
"Zhang",
"Li",
""
],
[
"Liu",
"Fang",
""
],
[
"Zhu",
"Yinghao",
""
],
[
"Xu",
"Wang",
""
],
[
"Shi",
"Lin",
""
],
[
"Lian",
"Xiaoli",
""
],
[
"Li",
"Minxiao",
""
],
[
"Shen",
"Bo",
""
],
[
"Fu",
"An",
""
]
] |
2506.02781 | Tongyuan Bai | Tongyuan Bai and Wangyuanfan Bai and Dong Chen and Tieru Wu and Manyi Li and Rui Ma | FreeScene: Mixed Graph Diffusion for 3D Scene Synthesis from Free Prompts | Accepted to CVPR 2025 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Controllability plays a crucial role in the practical applications of 3D indoor scene synthesis. Existing works either allow rough language-based control, that is convenient but lacks fine-grained scene customization, or employ graph based control, which offers better controllability but demands considerable knowledge for the cumbersome graph design process. To address these challenges, we present FreeScene, a user-friendly framework that enables both convenient and effective control for indoor scene synthesis.Specifically, FreeScene supports free-form user inputs including text description and/or reference images, allowing users to express versatile design intentions. The user inputs are adequately analyzed and integrated into a graph representation by a VLM-based Graph Designer. We then propose MG-DiT, a Mixed Graph Diffusion Transformer, which performs graph-aware denoising to enhance scene generation. Our MG-DiT not only excels at preserving graph structure but also offers broad applicability to various tasks, including, but not limited to, text-to-scene, graph-to-scene, and rearrangement, all within a single model. Extensive experiments demonstrate that FreeScene provides an efficient and user-friendly solution that unifies text-based and graph based scene synthesis, outperforming state-of-the-art methods in terms of both generation quality and controllability in a range of applications. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 12:01:41 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Bai",
"Tongyuan",
""
],
[
"Bai",
"Wangyuanfan",
""
],
[
"Chen",
"Dong",
""
],
[
"Wu",
"Tieru",
""
],
[
"Li",
"Manyi",
""
],
[
"Ma",
"Rui",
""
]
] |
2506.02782 | Medina Bandic | Hila Safi, Medina Bandic, Christoph Niedermeier, Carmen G. Almudever, Sebastian Feld, Wolfgang Mauerer | Stacking the Odds: Full-Stack Quantum System Design Space Exploration | null | null | null | null | quant-ph cs.ET | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Design space exploration (DSE) plays an important role in optimising quantum circuit execution by systematically evaluating different configurations of compilation strategies and hardware settings. In this work, we study the impact of layout methods, qubit routing techniques, compiler optimization levels, and hardware-specific properties, including noise characteristics, topological structures, connectivity densities, and device sizes. By traversing these dimensions, we aim to understand how compilation choices interact with hardware features. A central question in our study is whether carefully selected device parameters and mapping strategies, including initial layouts and routing heuristics, can mitigate hardware-induced errors beyond standard error mitigation methods. Our results show that choosing the right software strategies (e.g., layout and routing) and tailoring hardware properties (e.g., reducing noise or leveraging connectivity) significantly enhances the fidelity of quantum circuit executions. We provide performance estimates using metrics such as circuit depth, gate count, and expected fidelity. These findings highlight the value of hardware-software co-design, especially as quantum systems scale and move toward error-corrected computing. Our simulations, though noisy, include quantum error correction (QEC) scenarios, revealing similar sensitivities to layout and connectivity. This suggests that co-design principles will be vital for integrating QEC in future devices. Overall, we offer practical guidance for co-optimizing mapping, routing, and hardware configuration in real-world quantum computing. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 12:01:58 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Safi",
"Hila",
""
],
[
"Bandic",
"Medina",
""
],
[
"Niedermeier",
"Christoph",
""
],
[
"Almudever",
"Carmen G.",
""
],
[
"Feld",
"Sebastian",
""
],
[
"Mauerer",
"Wolfgang",
""
]
] |
2506.02783 | Carlos Garcia-Lopez-De-Haro | Carlos Garcia-Lopez-de-Haro, Caterina Fuster-Barcelo, Curtis T. Rueden, Jonathan Heras, Vladimir Ulman, Daniel Franco-Barranco, Adrian Ines, Kevin W. Eliceiri, Jean-Christophe Olivo-Marin, Jean-Yves Tinevez, Daniel Sage, Arrate Munoz-Barrutia | SAMJ: Fast Image Annotation on ImageJ/Fiji via Segment Anything Model | null | null | null | null | cs.CV | http://creativecommons.org/publicdomain/zero/1.0/ | Mask annotation remains a significant bottleneck in AI-driven biomedical image analysis due to its labor-intensive nature. To address this challenge, we introduce SAMJ, a user-friendly ImageJ/Fiji plugin leveraging the Segment Anything Model (SAM). SAMJ enables seamless, interactive annotations with one-click installation on standard computers. Designed for real-time object delineation in large scientific images, SAMJ is an easy-to-use solution that simplifies and accelerates the creation of labeled image datasets. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 12:10:03 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Garcia-Lopez-de-Haro",
"Carlos",
""
],
[
"Fuster-Barcelo",
"Caterina",
""
],
[
"Rueden",
"Curtis T.",
""
],
[
"Heras",
"Jonathan",
""
],
[
"Ulman",
"Vladimir",
""
],
[
"Franco-Barranco",
"Daniel",
""
],
[
"Ines",
"Adrian",
""
],
[
"Eliceiri",
"Kevin W.",
""
],
[
"Olivo-Marin",
"Jean-Christophe",
""
],
[
"Tinevez",
"Jean-Yves",
""
],
[
"Sage",
"Daniel",
""
],
[
"Munoz-Barrutia",
"Arrate",
""
]
] |
2506.02784 | Yue Zhang | Yue Zhang, Yankai Chen, Yingli Zhou, Yucan Guo, Xiaolin Han, Chenhao Ma | UTCS: Effective Unsupervised Temporal Community Search with Pre-training of Temporal Dynamics and Subgraph Knowledge | Accepted by SIGIR'25 short paper track | null | null | null | cs.IR | http://creativecommons.org/licenses/by/4.0/ | In many real-world applications, the evolving relationships between entities can be modeled as temporal graphs, where each edge has a timestamp representing the interaction time.
As a fundamental problem in graph analysis, {\it community search (CS)} in temporal graphs has received growing attention but exhibits two major limitations: (1) Traditional methods typically require predefined subgraph structures, which are not always known in advance. (2) Learning-based methods struggle to capture temporal interaction information. To fill this research gap, in this paper, we propose an effective \textbf{U}nsupervised \textbf{T}emporal \textbf{C}ommunity \textbf{S}earch with pre-training of temporal dynamics and subgraph knowledge model (\textbf{\model}). \model~contains two key stages: offline pre-training and online search. In the first stage, we introduce multiple learning objectives to facilitate the pre-training process in the unsupervised learning setting. In the second stage, we identify a candidate subgraph and compute community scores using the pre-trained node representations and a novel scoring mechanism to determine the final community members. Experiments on five real-world datasets demonstrate the effectiveness. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 12:11:34 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Zhang",
"Yue",
""
],
[
"Chen",
"Yankai",
""
],
[
"Zhou",
"Yingli",
""
],
[
"Guo",
"Yucan",
""
],
[
"Han",
"Xiaolin",
""
],
[
"Ma",
"Chenhao",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.