id
large_stringlengths 9
16
| submitter
large_stringlengths 3
64
⌀ | authors
large_stringlengths 5
20.2k
| title
large_stringlengths 5
245
| comments
large_stringlengths 1
482
⌀ | journal-ref
large_stringlengths 4
382
⌀ | doi
large_stringlengths 9
151
⌀ | report-no
large_stringlengths 2
509
⌀ | categories
large_stringlengths 5
78
| license
large_stringclasses 9
values | abstract
large_stringlengths 83
3.55k
| versions
large listlengths 1
21
| update_date
timestamp[ms]date 2007-05-23 00:00:00
2025-05-30 00:00:00
| authors_parsed
large listlengths 1
1.35k
| prompt
large_stringlengths 166
3.64k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2306.04101 | Xiusi Chen | Xiusi Chen, Yu Zhang, Jinliang Deng, Jyun-Yu Jiang, Wei Wang | Gotta: Generative Few-shot Question Answering by Prompt-based Cloze Data
Augmentation | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Few-shot question answering (QA) aims at precisely discovering answers to a
set of questions from context passages while only a few training samples are
available. Although existing studies have made some progress and can usually
achieve proper results, they suffer from understanding deep semantics for
reasoning out the questions. In this paper, we develop Gotta, a Generative
prOmpT-based daTa Augmentation framework to mitigate the challenge above.
Inspired by the human reasoning process, we propose to integrate the cloze task
to enhance few-shot QA learning. Following the recent success of prompt-tuning,
we present the cloze task in the same format as the main QA task, allowing the
model to learn both tasks seamlessly together to fully take advantage of the
power of prompt-tuning. Extensive experiments on widely used benchmarks
demonstrate that Gotta consistently outperforms competitive baselines,
validating the effectiveness of our proposed prompt-tuning-based cloze task,
which not only fine-tunes language models but also learns to guide reasoning in
QA tasks. Further analysis shows that the prompt-based loss incorporates the
auxiliary task better than the multi-task loss, highlighting the strength of
prompt-tuning on the few-shot QA task.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 01:44:43 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Chen",
"Xiusi",
""
],
[
"Zhang",
"Yu",
""
],
[
"Deng",
"Jinliang",
""
],
[
"Jiang",
"Jyun-Yu",
""
],
[
"Wang",
"Wei",
""
]
] | TITLE: Gotta: Generative Few-shot Question Answering by Prompt-based Cloze Data
Augmentation
ABSTRACT: Few-shot question answering (QA) aims at precisely discovering answers to a
set of questions from context passages while only a few training samples are
available. Although existing studies have made some progress and can usually
achieve proper results, they suffer from understanding deep semantics for
reasoning out the questions. In this paper, we develop Gotta, a Generative
prOmpT-based daTa Augmentation framework to mitigate the challenge above.
Inspired by the human reasoning process, we propose to integrate the cloze task
to enhance few-shot QA learning. Following the recent success of prompt-tuning,
we present the cloze task in the same format as the main QA task, allowing the
model to learn both tasks seamlessly together to fully take advantage of the
power of prompt-tuning. Extensive experiments on widely used benchmarks
demonstrate that Gotta consistently outperforms competitive baselines,
validating the effectiveness of our proposed prompt-tuning-based cloze task,
which not only fine-tunes language models but also learns to guide reasoning in
QA tasks. Further analysis shows that the prompt-based loss incorporates the
auxiliary task better than the multi-task loss, highlighting the strength of
prompt-tuning on the few-shot QA task.
|
2306.04109 | Jia Cheng Xu | JiaCheng Xu and ChengXiang Tan | Membership inference attack with relative decision boundary distance | 12 pages,9 figures, 5 tables | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Membership inference attack is one of the most popular privacy attacks in
machine learning, which aims to predict whether a given sample was contained in
the target model's training set. Label-only membership inference attack is a
variant that exploits sample robustness and attracts more attention since it
assumes a practical scenario in which the adversary only has access to the
predicted labels of the input samples. However, since the decision boundary
distance, which measures robustness, is strongly affected by the random initial
image, the adversary may get opposite results even for the same input samples.
In this paper, we propose a new attack method, called muti-class adaptive
membership inference attack in the label-only setting. All decision boundary
distances for all target classes have been traversed in the early attack
iterations, and the subsequent attack iterations continue with the shortest
decision boundary distance to obtain a stable and optimal decision boundary
distance. Instead of using a single boundary distance, the relative boundary
distance between samples and neighboring points has also been employed as a new
membership score to distinguish between member samples inside the training set
and nonmember samples outside the training set. Experiments show that previous
label-only membership inference attacks using the untargeted HopSkipJump
algorithm fail to achieve optimal decision bounds in more than half of the
samples, whereas our multi-targeted HopSkipJump algorithm succeeds in almost
all samples. In addition, extensive experiments show that our multi-class
adaptive MIA outperforms current label-only membership inference attacks in the
CIFAR10, and CIFAR100 datasets, especially for the true positive rate at low
false positive rates metric.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 02:29:58 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Xu",
"JiaCheng",
""
],
[
"Tan",
"ChengXiang",
""
]
] | TITLE: Membership inference attack with relative decision boundary distance
ABSTRACT: Membership inference attack is one of the most popular privacy attacks in
machine learning, which aims to predict whether a given sample was contained in
the target model's training set. Label-only membership inference attack is a
variant that exploits sample robustness and attracts more attention since it
assumes a practical scenario in which the adversary only has access to the
predicted labels of the input samples. However, since the decision boundary
distance, which measures robustness, is strongly affected by the random initial
image, the adversary may get opposite results even for the same input samples.
In this paper, we propose a new attack method, called muti-class adaptive
membership inference attack in the label-only setting. All decision boundary
distances for all target classes have been traversed in the early attack
iterations, and the subsequent attack iterations continue with the shortest
decision boundary distance to obtain a stable and optimal decision boundary
distance. Instead of using a single boundary distance, the relative boundary
distance between samples and neighboring points has also been employed as a new
membership score to distinguish between member samples inside the training set
and nonmember samples outside the training set. Experiments show that previous
label-only membership inference attacks using the untargeted HopSkipJump
algorithm fail to achieve optimal decision bounds in more than half of the
samples, whereas our multi-targeted HopSkipJump algorithm succeeds in almost
all samples. In addition, extensive experiments show that our multi-class
adaptive MIA outperforms current label-only membership inference attacks in the
CIFAR10, and CIFAR100 datasets, especially for the true positive rate at low
false positive rates metric.
|
2306.04116 | Yuki Arase | Yuki Arase, Han Bao, Sho Yokoi | Unbalanced Optimal Transport for Unbalanced Word Alignment | Accepted for the Annual Meeting of the Association for Computational
Linguistics (ACL 2023) | null | null | null | cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Monolingual word alignment is crucial to model semantic interactions between
sentences. In particular, null alignment, a phenomenon in which words have no
corresponding counterparts, is pervasive and critical in handling semantically
divergent sentences. Identification of null alignment is useful on its own to
reason about the semantic similarity of sentences by indicating there exists
information inequality. To achieve unbalanced word alignment that values both
alignment and null alignment, this study shows that the family of optimal
transport (OT), i.e., balanced, partial, and unbalanced OT, are natural and
powerful approaches even without tailor-made techniques. Our extensive
experiments covering unsupervised and supervised settings indicate that our
generic OT-based alignment methods are competitive against the
state-of-the-arts specially designed for word alignment, remarkably on
challenging datasets with high null alignment frequencies.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 03:03:41 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Arase",
"Yuki",
""
],
[
"Bao",
"Han",
""
],
[
"Yokoi",
"Sho",
""
]
] | TITLE: Unbalanced Optimal Transport for Unbalanced Word Alignment
ABSTRACT: Monolingual word alignment is crucial to model semantic interactions between
sentences. In particular, null alignment, a phenomenon in which words have no
corresponding counterparts, is pervasive and critical in handling semantically
divergent sentences. Identification of null alignment is useful on its own to
reason about the semantic similarity of sentences by indicating there exists
information inequality. To achieve unbalanced word alignment that values both
alignment and null alignment, this study shows that the family of optimal
transport (OT), i.e., balanced, partial, and unbalanced OT, are natural and
powerful approaches even without tailor-made techniques. Our extensive
experiments covering unsupervised and supervised settings indicate that our
generic OT-based alignment methods are competitive against the
state-of-the-arts specially designed for word alignment, remarkably on
challenging datasets with high null alignment frequencies.
|
2306.04118 | Yinghao Zhu | Yinghao Zhu, Jingkun An, Enshen Zhou, Lu An, Junyi Gao, Hao Li, Haoran
Feng, Bo Hou, Wen Tang, Chengwei Pan, Liantao Ma | M$^3$Fair: Mitigating Bias in Healthcare Data through Multi-Level and
Multi-Sensitive-Attribute Reweighting Method | 4 pages, 1 table, Beijing Health Data Science Summit 2023 | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | In the data-driven artificial intelligence paradigm, models heavily rely on
large amounts of training data. However, factors like sampling distribution
imbalance can lead to issues of bias and unfairness in healthcare data.
Sensitive attributes, such as race, gender, age, and medical condition, are
characteristics of individuals that are commonly associated with discrimination
or bias. In healthcare AI, these attributes can play a significant role in
determining the quality of care that individuals receive. For example, minority
groups often receive fewer procedures and poorer-quality medical care than
white individuals in US. Therefore, detecting and mitigating bias in data is
crucial to enhancing health equity. Bias mitigation methods include
pre-processing, in-processing, and post-processing. Among them, Reweighting
(RW) is a widely used pre-processing method that performs well in balancing
machine learning performance and fairness performance. RW adjusts the weights
for samples within each (group, label) combination, where these weights are
utilized in loss functions. However, RW is limited to considering only a single
sensitive attribute when mitigating bias and assumes that each sensitive
attribute is equally important. This may result in potential inaccuracies when
addressing intersectional bias. To address these limitations, we propose
M3Fair, a multi-level and multi-sensitive-attribute reweighting method by
extending the RW method to multiple sensitive attributes at multiple levels.
Our experiments on real-world datasets show that the approach is effective,
straightforward, and generalizable in addressing the healthcare fairness
issues.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 03:20:44 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Zhu",
"Yinghao",
""
],
[
"An",
"Jingkun",
""
],
[
"Zhou",
"Enshen",
""
],
[
"An",
"Lu",
""
],
[
"Gao",
"Junyi",
""
],
[
"Li",
"Hao",
""
],
[
"Feng",
"Haoran",
""
],
[
"Hou",
"Bo",
""
],
[
"Tang",
"Wen",
""
],
[
"Pan",
"Chengwei",
""
],
[
"Ma",
"Liantao",
""
]
] | TITLE: M$^3$Fair: Mitigating Bias in Healthcare Data through Multi-Level and
Multi-Sensitive-Attribute Reweighting Method
ABSTRACT: In the data-driven artificial intelligence paradigm, models heavily rely on
large amounts of training data. However, factors like sampling distribution
imbalance can lead to issues of bias and unfairness in healthcare data.
Sensitive attributes, such as race, gender, age, and medical condition, are
characteristics of individuals that are commonly associated with discrimination
or bias. In healthcare AI, these attributes can play a significant role in
determining the quality of care that individuals receive. For example, minority
groups often receive fewer procedures and poorer-quality medical care than
white individuals in US. Therefore, detecting and mitigating bias in data is
crucial to enhancing health equity. Bias mitigation methods include
pre-processing, in-processing, and post-processing. Among them, Reweighting
(RW) is a widely used pre-processing method that performs well in balancing
machine learning performance and fairness performance. RW adjusts the weights
for samples within each (group, label) combination, where these weights are
utilized in loss functions. However, RW is limited to considering only a single
sensitive attribute when mitigating bias and assumes that each sensitive
attribute is equally important. This may result in potential inaccuracies when
addressing intersectional bias. To address these limitations, we propose
M3Fair, a multi-level and multi-sensitive-attribute reweighting method by
extending the RW method to multiple sensitive attributes at multiple levels.
Our experiments on real-world datasets show that the approach is effective,
straightforward, and generalizable in addressing the healthcare fairness
issues.
|
2306.04123 | Shufang Xie | Shufang Xie, Rui Yan, Junliang Guo, Yingce Xia, Lijun Wu, Tao Qin | Retrosynthesis Prediction with Local Template Retrieval | AAAI-2023 camera ready | null | null | null | cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Retrosynthesis, which predicts the reactants of a given target molecule, is
an essential task for drug discovery. In recent years, the machine learing
based retrosynthesis methods have achieved promising results. In this work, we
introduce RetroKNN, a local reaction template retrieval method to further boost
the performance of template-based systems with non-parametric retrieval. We
first build an atom-template store and a bond-template store that contain the
local templates in the training data, then retrieve from these templates with a
k-nearest-neighbor (KNN) search during inference. The retrieved templates are
combined with neural network predictions as the final output. Furthermore, we
propose a lightweight adapter to adjust the weights when combing neural network
and KNN predictions conditioned on the hidden representation and the retrieved
templates. We conduct comprehensive experiments on two widely used benchmarks,
the USPTO-50K and USPTO-MIT. Especially for the top-1 accuracy, we improved
7.1% on the USPTO-50K dataset and 12.0% on the USPTO-MIT dataset. These results
demonstrate the effectiveness of our method.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 03:38:03 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Xie",
"Shufang",
""
],
[
"Yan",
"Rui",
""
],
[
"Guo",
"Junliang",
""
],
[
"Xia",
"Yingce",
""
],
[
"Wu",
"Lijun",
""
],
[
"Qin",
"Tao",
""
]
] | TITLE: Retrosynthesis Prediction with Local Template Retrieval
ABSTRACT: Retrosynthesis, which predicts the reactants of a given target molecule, is
an essential task for drug discovery. In recent years, the machine learing
based retrosynthesis methods have achieved promising results. In this work, we
introduce RetroKNN, a local reaction template retrieval method to further boost
the performance of template-based systems with non-parametric retrieval. We
first build an atom-template store and a bond-template store that contain the
local templates in the training data, then retrieve from these templates with a
k-nearest-neighbor (KNN) search during inference. The retrieved templates are
combined with neural network predictions as the final output. Furthermore, we
propose a lightweight adapter to adjust the weights when combing neural network
and KNN predictions conditioned on the hidden representation and the retrieved
templates. We conduct comprehensive experiments on two widely used benchmarks,
the USPTO-50K and USPTO-MIT. Especially for the top-1 accuracy, we improved
7.1% on the USPTO-50K dataset and 12.0% on the USPTO-MIT dataset. These results
demonstrate the effectiveness of our method.
|
2306.04133 | Shib Sankar Dasgupta | Shib Dasgupta, Andrew McCallum, Steffen Rendle, Li Zhang | Answering Compositional Queries with Set-Theoretic Embeddings | null | null | null | null | cs.IR cs.LG | http://creativecommons.org/licenses/by/4.0/ | The need to compactly and robustly represent item-attribute relations arises
in many important tasks, such as faceted browsing and recommendation systems. A
popular machine learning approach for this task denotes that an item has an
attribute by a high dot-product between vectors for the item and attribute -- a
representation that is not only dense, but also tends to correct noisy and
incomplete data. While this method works well for queries retrieving items by a
single attribute (such as \emph{movies that are comedies}), we find that vector
embeddings do not so accurately support compositional queries (such as movies
that are comedies and British but not romances). To address these set-theoretic
compositions, this paper proposes to replace vectors with box embeddings, a
region-based representation that can be thought of as learnable Venn diagrams.
We introduce a new benchmark dataset for compositional queries, and present
experiments and analysis providing insights into the behavior of both. We find
that, while vector and box embeddings are equally suited to single attribute
queries, for compositional queries box embeddings provide substantial
advantages over vectors, particularly at the moderate and larger retrieval set
sizes that are most useful for users' search and browsing.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 04:04:36 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Dasgupta",
"Shib",
""
],
[
"McCallum",
"Andrew",
""
],
[
"Rendle",
"Steffen",
""
],
[
"Zhang",
"Li",
""
]
] | TITLE: Answering Compositional Queries with Set-Theoretic Embeddings
ABSTRACT: The need to compactly and robustly represent item-attribute relations arises
in many important tasks, such as faceted browsing and recommendation systems. A
popular machine learning approach for this task denotes that an item has an
attribute by a high dot-product between vectors for the item and attribute -- a
representation that is not only dense, but also tends to correct noisy and
incomplete data. While this method works well for queries retrieving items by a
single attribute (such as \emph{movies that are comedies}), we find that vector
embeddings do not so accurately support compositional queries (such as movies
that are comedies and British but not romances). To address these set-theoretic
compositions, this paper proposes to replace vectors with box embeddings, a
region-based representation that can be thought of as learnable Venn diagrams.
We introduce a new benchmark dataset for compositional queries, and present
experiments and analysis providing insights into the behavior of both. We find
that, while vector and box embeddings are equally suited to single attribute
queries, for compositional queries box embeddings provide substantial
advantages over vectors, particularly at the moderate and larger retrieval set
sizes that are most useful for users' search and browsing.
|
2306.04148 | Chandan Misra | Chandan Misra and Swarup Chattopadhyay | SANGEET: A XML based Open Dataset for Research in Hindustani Sangeet | null | null | null | null | cs.SD cs.IR cs.LG eess.AS | http://creativecommons.org/licenses/by/4.0/ | It is very important to access a rich music dataset that is useful in a wide
variety of applications. Currently, available datasets are mostly focused on
storing vocal or instrumental recording data and ignoring the requirement of
its visual representation and retrieval. This paper attempts to build an
XML-based public dataset, called SANGEET, that stores comprehensive information
of Hindustani Sangeet (North Indian Classical Music) compositions written by
famous musicologist Pt. Vishnu Narayan Bhatkhande. SANGEET preserves all the
required information of any given composition including metadata, structural,
notational, rhythmic, and melodic information in a standardized way for easy
and efficient storage and extraction of musical information. The dataset is
intended to provide the ground truth information for music information research
tasks, thereby supporting several data-driven analysis from a machine learning
perspective. We present the usefulness of the dataset by demonstrating its
application on music information retrieval using XQuery, visualization through
Omenad rendering system. Finally, we propose approaches to transform the
dataset for performing statistical and machine learning tasks for a better
understanding of Hindustani Sangeet. The dataset can be found at
https://github.com/cmisra/Sangeet.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 04:50:09 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Misra",
"Chandan",
""
],
[
"Chattopadhyay",
"Swarup",
""
]
] | TITLE: SANGEET: A XML based Open Dataset for Research in Hindustani Sangeet
ABSTRACT: It is very important to access a rich music dataset that is useful in a wide
variety of applications. Currently, available datasets are mostly focused on
storing vocal or instrumental recording data and ignoring the requirement of
its visual representation and retrieval. This paper attempts to build an
XML-based public dataset, called SANGEET, that stores comprehensive information
of Hindustani Sangeet (North Indian Classical Music) compositions written by
famous musicologist Pt. Vishnu Narayan Bhatkhande. SANGEET preserves all the
required information of any given composition including metadata, structural,
notational, rhythmic, and melodic information in a standardized way for easy
and efficient storage and extraction of musical information. The dataset is
intended to provide the ground truth information for music information research
tasks, thereby supporting several data-driven analysis from a machine learning
perspective. We present the usefulness of the dataset by demonstrating its
application on music information retrieval using XQuery, visualization through
Omenad rendering system. Finally, we propose approaches to transform the
dataset for performing statistical and machine learning tasks for a better
understanding of Hindustani Sangeet. The dataset can be found at
https://github.com/cmisra/Sangeet.
|
2306.04152 | Haiqin Yang | Junxian Zhou, Haiqin Yang, Yuxuan He, Hao Mou, Junbo Yang | A Unified One-Step Solution for Aspect Sentiment Quad Prediction | 15 pages, 12 tables, 3 figures, ACL Findings | null | null | null | cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Aspect sentiment quad prediction (ASQP) is a challenging yet significant
subtask in aspect-based sentiment analysis as it provides a complete
aspect-level sentiment structure. However, existing ASQP datasets are usually
small and low-density, hindering technical advancement. To expand the capacity,
in this paper, we release two new datasets for ASQP, which contain the
following characteristics: larger size, more words per sample, and higher
density. With such datasets, we unveil the shortcomings of existing strong ASQP
baselines and therefore propose a unified one-step solution for ASQP, namely
One-ASQP, to detect the aspect categories and to identify the
aspect-opinion-sentiment (AOS) triplets simultaneously. Our One-ASQP holds
several unique advantages: (1) by separating ASQP into two subtasks and solving
them independently and simultaneously, we can avoid error propagation in
pipeline-based methods and overcome slow training and inference in
generation-based methods; (2) by introducing sentiment-specific horns tagging
schema in a token-pair-based two-dimensional matrix, we can exploit deeper
interactions between sentiment elements and efficiently decode the AOS
triplets; (3) we design ``[NULL]'' token can help us effectively identify the
implicit aspects or opinions. Experiments on two benchmark datasets and our
released two datasets demonstrate the advantages of our One-ASQP. The two new
datasets are publicly released at
\url{https://www.github.com/Datastory-CN/ASQP-Datasets}.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 05:00:01 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Zhou",
"Junxian",
""
],
[
"Yang",
"Haiqin",
""
],
[
"He",
"Yuxuan",
""
],
[
"Mou",
"Hao",
""
],
[
"Yang",
"Junbo",
""
]
] | TITLE: A Unified One-Step Solution for Aspect Sentiment Quad Prediction
ABSTRACT: Aspect sentiment quad prediction (ASQP) is a challenging yet significant
subtask in aspect-based sentiment analysis as it provides a complete
aspect-level sentiment structure. However, existing ASQP datasets are usually
small and low-density, hindering technical advancement. To expand the capacity,
in this paper, we release two new datasets for ASQP, which contain the
following characteristics: larger size, more words per sample, and higher
density. With such datasets, we unveil the shortcomings of existing strong ASQP
baselines and therefore propose a unified one-step solution for ASQP, namely
One-ASQP, to detect the aspect categories and to identify the
aspect-opinion-sentiment (AOS) triplets simultaneously. Our One-ASQP holds
several unique advantages: (1) by separating ASQP into two subtasks and solving
them independently and simultaneously, we can avoid error propagation in
pipeline-based methods and overcome slow training and inference in
generation-based methods; (2) by introducing sentiment-specific horns tagging
schema in a token-pair-based two-dimensional matrix, we can exploit deeper
interactions between sentiment elements and efficiently decode the AOS
triplets; (3) we design ``[NULL]'' token can help us effectively identify the
implicit aspects or opinions. Experiments on two benchmark datasets and our
released two datasets demonstrate the advantages of our One-ASQP. The two new
datasets are publicly released at
\url{https://www.github.com/Datastory-CN/ASQP-Datasets}.
|
2306.04163 | Mengyu Chen | Mengyu Chen, Zhenchang Xing, Jieshan Chen, Chunyang Chen and Qinghua
Lu | Enhancing Virtual Assistant Intelligence: Precise Area Targeting for
Instance-level User Intents beyond Metadata | null | null | null | null | cs.HC cs.AI cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Virtual assistants have been widely used by mobile phone users in recent
years. Although their capabilities of processing user intents have been
developed rapidly, virtual assistants in most platforms are only capable of
handling pre-defined high-level tasks supported by extra manual efforts of
developers. However, instance-level user intents containing more detailed
objectives with complex practical situations, are yet rarely studied so far. In
this paper, we explore virtual assistants capable of processing instance-level
user intents based on pixels of application screens, without the requirements
of extra extensions on the application side. We propose a novel cross-modal
deep learning pipeline, which understands the input vocal or textual
instance-level user intents, predicts the targeting operational area, and
detects the absolute button area on screens without any metadata of
applications. We conducted a user study with 10 participants to collect a
testing dataset with instance-level user intents. The testing dataset is then
utilized to evaluate the performance of our model, which demonstrates that our
model is promising with the achievement of 64.43% accuracy on our testing
dataset.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 05:26:38 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Chen",
"Mengyu",
""
],
[
"Xing",
"Zhenchang",
""
],
[
"Chen",
"Jieshan",
""
],
[
"Chen",
"Chunyang",
""
],
[
"Lu",
"Qinghua",
""
]
] | TITLE: Enhancing Virtual Assistant Intelligence: Precise Area Targeting for
Instance-level User Intents beyond Metadata
ABSTRACT: Virtual assistants have been widely used by mobile phone users in recent
years. Although their capabilities of processing user intents have been
developed rapidly, virtual assistants in most platforms are only capable of
handling pre-defined high-level tasks supported by extra manual efforts of
developers. However, instance-level user intents containing more detailed
objectives with complex practical situations, are yet rarely studied so far. In
this paper, we explore virtual assistants capable of processing instance-level
user intents based on pixels of application screens, without the requirements
of extra extensions on the application side. We propose a novel cross-modal
deep learning pipeline, which understands the input vocal or textual
instance-level user intents, predicts the targeting operational area, and
detects the absolute button area on screens without any metadata of
applications. We conducted a user study with 10 participants to collect a
testing dataset with instance-level user intents. The testing dataset is then
utilized to evaluate the performance of our model, which demonstrates that our
model is promising with the achievement of 64.43% accuracy on our testing
dataset.
|
2306.04170 | Zhibin Chen | Zhibin Chen, Yansong Feng, Dongyan Zhao | From the One, Judge of the Whole: Typed Entailment Graph Construction
with Predicate Generation | 9 pages, 3 figures, accepted to ACL 2023 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Entailment Graphs (EGs) have been constructed based on extracted corpora as a
strong and explainable form to indicate context-independent entailment
relations in natural languages. However, EGs built by previous methods often
suffer from the severe sparsity issues, due to limited corpora available and
the long-tail phenomenon of predicate distributions. In this paper, we propose
a multi-stage method, Typed Predicate-Entailment Graph Generator (TP-EGG), to
tackle this problem. Given several seed predicates, TP-EGG builds the graphs by
generating new predicates and detecting entailment relations among them. The
generative nature of TP-EGG helps us leverage the recent advances from large
pretrained language models (PLMs), while avoiding the reliance on carefully
prepared corpora. Experiments on benchmark datasets show that TP-EGG can
generate high-quality and scale-controllable entailment graphs, achieving
significant in-domain improvement over state-of-the-art EGs and boosting the
performance of down-stream inference tasks.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 05:46:19 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Chen",
"Zhibin",
""
],
[
"Feng",
"Yansong",
""
],
[
"Zhao",
"Dongyan",
""
]
] | TITLE: From the One, Judge of the Whole: Typed Entailment Graph Construction
with Predicate Generation
ABSTRACT: Entailment Graphs (EGs) have been constructed based on extracted corpora as a
strong and explainable form to indicate context-independent entailment
relations in natural languages. However, EGs built by previous methods often
suffer from the severe sparsity issues, due to limited corpora available and
the long-tail phenomenon of predicate distributions. In this paper, we propose
a multi-stage method, Typed Predicate-Entailment Graph Generator (TP-EGG), to
tackle this problem. Given several seed predicates, TP-EGG builds the graphs by
generating new predicates and detecting entailment relations among them. The
generative nature of TP-EGG helps us leverage the recent advances from large
pretrained language models (PLMs), while avoiding the reliance on carefully
prepared corpora. Experiments on benchmark datasets show that TP-EGG can
generate high-quality and scale-controllable entailment graphs, achieving
significant in-domain improvement over state-of-the-art EGs and boosting the
performance of down-stream inference tasks.
|
2306.04176 | Sang-Eun Han | Kyungjae Lee, Sang-eun Han, Seung-won Hwang, Moontae Lee | When to Read Documents or QA History: On Unified and Selective
Open-domain QA | Findings of ACL 2023 camera ready | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper studies the problem of open-domain question answering, with the
aim of answering a diverse range of questions leveraging knowledge resources.
Two types of sources, QA-pair and document corpora, have been actively
leveraged with the following complementary strength. The former is highly
precise when the paraphrase of given question $q$ was seen and answered during
training, often posed as a retrieval problem, while the latter generalizes
better for unseen questions. A natural follow-up is thus leveraging both
models, while a naive pipelining or integration approaches have failed to bring
additional gains over either model alone. Our distinction is interpreting the
problem as calibration, which estimates the confidence of predicted answers as
an indicator to decide when to use a document or QA-pair corpus. The
effectiveness of our method was validated on widely adopted benchmarks such as
Natural Questions and TriviaQA.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 06:03:39 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Lee",
"Kyungjae",
""
],
[
"Han",
"Sang-eun",
""
],
[
"Hwang",
"Seung-won",
""
],
[
"Lee",
"Moontae",
""
]
] | TITLE: When to Read Documents or QA History: On Unified and Selective
Open-domain QA
ABSTRACT: This paper studies the problem of open-domain question answering, with the
aim of answering a diverse range of questions leveraging knowledge resources.
Two types of sources, QA-pair and document corpora, have been actively
leveraged with the following complementary strength. The former is highly
precise when the paraphrase of given question $q$ was seen and answered during
training, often posed as a retrieval problem, while the latter generalizes
better for unseen questions. A natural follow-up is thus leveraging both
models, while a naive pipelining or integration approaches have failed to bring
additional gains over either model alone. Our distinction is interpreting the
problem as calibration, which estimates the confidence of predicted answers as
an indicator to decide when to use a document or QA-pair corpus. The
effectiveness of our method was validated on widely adopted benchmarks such as
Natural Questions and TriviaQA.
|
2306.04184 | Han Hu | Libin Wang, Han Hu, Qisen Shang, Bo Xu, Qing Zhu | StructuredMesh: 3D Structured Optimization of Fa\c{c}ade Components on
Photogrammetric Mesh Models using Binary Integer Programming | 30 pages,15 figures | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The lack of fa\c{c}ade structures in photogrammetric mesh models renders them
inadequate for meeting the demands of intricate applications. Moreover, these
mesh models exhibit irregular surfaces with considerable geometric noise and
texture quality imperfections, making the restoration of structures
challenging. To address these shortcomings, we present StructuredMesh, a novel
approach for reconstructing fa\c{c}ade structures conforming to the regularity
of buildings within photogrammetric mesh models. Our method involves capturing
multi-view color and depth images of the building model using a virtual camera
and employing a deep learning object detection pipeline to semi-automatically
extract the bounding boxes of fa\c{c}ade components such as windows, doors, and
balconies from the color image. We then utilize the depth image to remap these
boxes into 3D space, generating an initial fa\c{c}ade layout. Leveraging
architectural knowledge, we apply binary integer programming (BIP) to optimize
the 3D layout's structure, encompassing the positions, orientations, and sizes
of all components. The refined layout subsequently informs fa\c{c}ade modeling
through instance replacement. We conducted experiments utilizing building mesh
models from three distinct datasets, demonstrating the adaptability,
robustness, and noise resistance of our proposed methodology. Furthermore, our
3D layout evaluation metrics reveal that the optimized layout enhances
precision, recall, and F-score by 6.5%, 4.5%, and 5.5%, respectively, in
comparison to the initial layout.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 06:40:54 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Wang",
"Libin",
""
],
[
"Hu",
"Han",
""
],
[
"Shang",
"Qisen",
""
],
[
"Xu",
"Bo",
""
],
[
"Zhu",
"Qing",
""
]
] | TITLE: StructuredMesh: 3D Structured Optimization of Fa\c{c}ade Components on
Photogrammetric Mesh Models using Binary Integer Programming
ABSTRACT: The lack of fa\c{c}ade structures in photogrammetric mesh models renders them
inadequate for meeting the demands of intricate applications. Moreover, these
mesh models exhibit irregular surfaces with considerable geometric noise and
texture quality imperfections, making the restoration of structures
challenging. To address these shortcomings, we present StructuredMesh, a novel
approach for reconstructing fa\c{c}ade structures conforming to the regularity
of buildings within photogrammetric mesh models. Our method involves capturing
multi-view color and depth images of the building model using a virtual camera
and employing a deep learning object detection pipeline to semi-automatically
extract the bounding boxes of fa\c{c}ade components such as windows, doors, and
balconies from the color image. We then utilize the depth image to remap these
boxes into 3D space, generating an initial fa\c{c}ade layout. Leveraging
architectural knowledge, we apply binary integer programming (BIP) to optimize
the 3D layout's structure, encompassing the positions, orientations, and sizes
of all components. The refined layout subsequently informs fa\c{c}ade modeling
through instance replacement. We conducted experiments utilizing building mesh
models from three distinct datasets, demonstrating the adaptability,
robustness, and noise resistance of our proposed methodology. Furthermore, our
3D layout evaluation metrics reveal that the optimized layout enhances
precision, recall, and F-score by 6.5%, 4.5%, and 5.5%, respectively, in
comparison to the initial layout.
|
2306.04188 | Shiping Yang | Shiping Yang and Renliang Sun and Xiaojun Wan | A New Dataset and Empirical Study for Sentence Simplification in Chinese | Accepted by ACL2023 main conference | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Sentence Simplification is a valuable technique that can benefit language
learners and children a lot. However, current research focuses more on English
sentence simplification. The development of Chinese sentence simplification is
relatively slow due to the lack of data. To alleviate this limitation, this
paper introduces CSS, a new dataset for assessing sentence simplification in
Chinese. We collect manual simplifications from human annotators and perform
data analysis to show the difference between English and Chinese sentence
simplifications. Furthermore, we test several unsupervised and zero/few-shot
learning methods on CSS and analyze the automatic evaluation and human
evaluation results. In the end, we explore whether Large Language Models can
serve as high-quality Chinese sentence simplification systems by evaluating
them on CSS.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 06:47:34 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Yang",
"Shiping",
""
],
[
"Sun",
"Renliang",
""
],
[
"Wan",
"Xiaojun",
""
]
] | TITLE: A New Dataset and Empirical Study for Sentence Simplification in Chinese
ABSTRACT: Sentence Simplification is a valuable technique that can benefit language
learners and children a lot. However, current research focuses more on English
sentence simplification. The development of Chinese sentence simplification is
relatively slow due to the lack of data. To alleviate this limitation, this
paper introduces CSS, a new dataset for assessing sentence simplification in
Chinese. We collect manual simplifications from human annotators and perform
data analysis to show the difference between English and Chinese sentence
simplifications. Furthermore, we test several unsupervised and zero/few-shot
learning methods on CSS and analyze the automatic evaluation and human
evaluation results. In the end, we explore whether Large Language Models can
serve as high-quality Chinese sentence simplification systems by evaluating
them on CSS.
|
2306.04214 | Jianpeng Liao | Jianpeng Liao, Jun Yan and Qian Tao | DualHGNN: A Dual Hypergraph Neural Network for Semi-Supervised Node
Classification based on Multi-View Learning and Density Awareness | This work has been accepted by 2023 International Joint Conference on
Neural Networks (IJCNN 2023). arXiv admin note: text overlap with
arXiv:2201.11511 | null | null | null | cs.LG cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Graph-based semi-supervised node classification has been shown to become a
state-of-the-art approach in many applications with high research value and
significance. Most existing methods are only based on the original intrinsic or
artificially established graph structure which may not accurately reflect the
"true" correlation among data and are not optimal for semi-supervised node
classification in the downstream graph neural networks. Besides, while existing
graph-based methods mostly utilize the explicit graph structure, some implicit
information, for example, the density information, can also provide latent
information that can be further exploited. To address these limitations, this
paper proposes the Dual Hypergraph Neural Network (DualHGNN), a new dual
connection model integrating both hypergraph structure learning and hypergraph
representation learning simultaneously in a unified architecture. The DualHGNN
first leverages a multi-view hypergraph learning network to explore the optimal
hypergraph structure from multiple views, constrained by a consistency loss
proposed to improve its generalization. Then, DualHGNN employs a density-aware
hypergraph attention network to explore the high-order semantic correlation
among data points based on the density-aware attention mechanism. Extensive
experiments are conducted in various benchmark datasets, and the results
demonstrate the effectiveness of the proposed approach.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 07:40:04 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Liao",
"Jianpeng",
""
],
[
"Yan",
"Jun",
""
],
[
"Tao",
"Qian",
""
]
] | TITLE: DualHGNN: A Dual Hypergraph Neural Network for Semi-Supervised Node
Classification based on Multi-View Learning and Density Awareness
ABSTRACT: Graph-based semi-supervised node classification has been shown to become a
state-of-the-art approach in many applications with high research value and
significance. Most existing methods are only based on the original intrinsic or
artificially established graph structure which may not accurately reflect the
"true" correlation among data and are not optimal for semi-supervised node
classification in the downstream graph neural networks. Besides, while existing
graph-based methods mostly utilize the explicit graph structure, some implicit
information, for example, the density information, can also provide latent
information that can be further exploited. To address these limitations, this
paper proposes the Dual Hypergraph Neural Network (DualHGNN), a new dual
connection model integrating both hypergraph structure learning and hypergraph
representation learning simultaneously in a unified architecture. The DualHGNN
first leverages a multi-view hypergraph learning network to explore the optimal
hypergraph structure from multiple views, constrained by a consistency loss
proposed to improve its generalization. Then, DualHGNN employs a density-aware
hypergraph attention network to explore the high-order semantic correlation
among data points based on the density-aware attention mechanism. Extensive
experiments are conducted in various benchmark datasets, and the results
demonstrate the effectiveness of the proposed approach.
|
2306.04217 | Xiaobao Wu | Xiaobao Wu, Xinshuai Dong, Thong Nguyen, Anh Tuan Luu | Effective Neural Topic Modeling with Embedding Clustering Regularization | Accepted to ICML 2023 conference | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Topic models have been prevalent for decades with various applications.
However, existing topic models commonly suffer from the notorious topic
collapsing: discovered topics semantically collapse towards each other, leading
to highly repetitive topics, insufficient topic discovery, and damaged model
interpretability. In this paper, we propose a new neural topic model, Embedding
Clustering Regularization Topic Model (ECRTM). Besides the existing
reconstruction error, we propose a novel Embedding Clustering Regularization
(ECR), which forces each topic embedding to be the center of a separately
aggregated word embedding cluster in the semantic space. This enables each
produced topic to contain distinct word semantics, which alleviates topic
collapsing. Regularized by ECR, our ECRTM generates diverse and coherent topics
together with high-quality topic distributions of documents. Extensive
experiments on benchmark datasets demonstrate that ECRTM effectively addresses
the topic collapsing issue and consistently surpasses state-of-the-art
baselines in terms of topic quality, topic distributions of documents, and
downstream classification tasks.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 07:45:38 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Wu",
"Xiaobao",
""
],
[
"Dong",
"Xinshuai",
""
],
[
"Nguyen",
"Thong",
""
],
[
"Luu",
"Anh Tuan",
""
]
] | TITLE: Effective Neural Topic Modeling with Embedding Clustering Regularization
ABSTRACT: Topic models have been prevalent for decades with various applications.
However, existing topic models commonly suffer from the notorious topic
collapsing: discovered topics semantically collapse towards each other, leading
to highly repetitive topics, insufficient topic discovery, and damaged model
interpretability. In this paper, we propose a new neural topic model, Embedding
Clustering Regularization Topic Model (ECRTM). Besides the existing
reconstruction error, we propose a novel Embedding Clustering Regularization
(ECR), which forces each topic embedding to be the center of a separately
aggregated word embedding cluster in the semantic space. This enables each
produced topic to contain distinct word semantics, which alleviates topic
collapsing. Regularized by ECR, our ECRTM generates diverse and coherent topics
together with high-quality topic distributions of documents. Extensive
experiments on benchmark datasets demonstrate that ECRTM effectively addresses
the topic collapsing issue and consistently surpasses state-of-the-art
baselines in terms of topic quality, topic distributions of documents, and
downstream classification tasks.
|
2306.04234 | Xianyu Chen | Xianyu Chen, Jian Shen, Wei Xia, Jiarui Jin, Yakun Song, Weinan Zhang,
Weiwen Liu, Menghui Zhu, Ruiming Tang, Kai Dong, Dingyin Xia, Yong Yu | Set-to-Sequence Ranking-based Concept-aware Learning Path Recommendation | null | null | null | null | cs.IR cs.CY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | With the development of the online education system, personalized education
recommendation has played an essential role. In this paper, we focus on
developing path recommendation systems that aim to generating and recommending
an entire learning path to the given user in each session. Noticing that
existing approaches fail to consider the correlations of concepts in the path,
we propose a novel framework named Set-to-Sequence Ranking-based Concept-aware
Learning Path Recommendation (SRC), which formulates the recommendation task
under a set-to-sequence paradigm. Specifically, we first design a concept-aware
encoder module which can capture the correlations among the input learning
concepts. The outputs are then fed into a decoder module that sequentially
generates a path through an attention mechanism that handles correlations
between the learning and target concepts. Our recommendation policy is
optimized by policy gradient. In addition, we also introduce an auxiliary
module based on knowledge tracing to enhance the model's stability by
evaluating students' learning effects on learning concepts. We conduct
extensive experiments on two real-world public datasets and one industrial
dataset, and the experimental results demonstrate the superiority and
effectiveness of SRC. Code will be available at
https://gitee.com/mindspore/models/tree/master/research/recommend/SRC.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 08:24:44 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Chen",
"Xianyu",
""
],
[
"Shen",
"Jian",
""
],
[
"Xia",
"Wei",
""
],
[
"Jin",
"Jiarui",
""
],
[
"Song",
"Yakun",
""
],
[
"Zhang",
"Weinan",
""
],
[
"Liu",
"Weiwen",
""
],
[
"Zhu",
"Menghui",
""
],
[
"Tang",
"Ruiming",
""
],
[
"Dong",
"Kai",
""
],
[
"Xia",
"Dingyin",
""
],
[
"Yu",
"Yong",
""
]
] | TITLE: Set-to-Sequence Ranking-based Concept-aware Learning Path Recommendation
ABSTRACT: With the development of the online education system, personalized education
recommendation has played an essential role. In this paper, we focus on
developing path recommendation systems that aim to generating and recommending
an entire learning path to the given user in each session. Noticing that
existing approaches fail to consider the correlations of concepts in the path,
we propose a novel framework named Set-to-Sequence Ranking-based Concept-aware
Learning Path Recommendation (SRC), which formulates the recommendation task
under a set-to-sequence paradigm. Specifically, we first design a concept-aware
encoder module which can capture the correlations among the input learning
concepts. The outputs are then fed into a decoder module that sequentially
generates a path through an attention mechanism that handles correlations
between the learning and target concepts. Our recommendation policy is
optimized by policy gradient. In addition, we also introduce an auxiliary
module based on knowledge tracing to enhance the model's stability by
evaluating students' learning effects on learning concepts. We conduct
extensive experiments on two real-world public datasets and one industrial
dataset, and the experimental results demonstrate the superiority and
effectiveness of SRC. Code will be available at
https://gitee.com/mindspore/models/tree/master/research/recommend/SRC.
|
2306.04240 | Feiyang Han | Feiyang Han, Yun Miao, Zhaoyi Sun, Yimin Wei | T-ADAF: Adaptive Data Augmentation Framework for Image Classification
Network based on Tensor T-product Operator | null | null | null | null | cs.CV cs.NA math.NA | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Image classification is one of the most fundamental tasks in Computer Vision.
In practical applications, the datasets are usually not as abundant as those in
the laboratory and simulation, which is always called as Data Hungry. How to
extract the information of data more completely and effectively is very
important. Therefore, an Adaptive Data Augmentation Framework based on the
tensor T-product Operator is proposed in this paper, to triple one image data
to be trained and gain the result from all these three images together with
only less than 0.1% increase in the number of parameters. At the same time,
this framework serves the functions of column image embedding and global
feature intersection, enabling the model to obtain information in not only
spatial but frequency domain, and thus improving the prediction accuracy of the
model. Numerical experiments have been designed for several models, and the
results demonstrate the effectiveness of this adaptive framework. Numerical
experiments show that our data augmentation framework can improve the
performance of original neural network model by 2%, which provides competitive
results to state-of-the-art methods.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 08:30:44 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Han",
"Feiyang",
""
],
[
"Miao",
"Yun",
""
],
[
"Sun",
"Zhaoyi",
""
],
[
"Wei",
"Yimin",
""
]
] | TITLE: T-ADAF: Adaptive Data Augmentation Framework for Image Classification
Network based on Tensor T-product Operator
ABSTRACT: Image classification is one of the most fundamental tasks in Computer Vision.
In practical applications, the datasets are usually not as abundant as those in
the laboratory and simulation, which is always called as Data Hungry. How to
extract the information of data more completely and effectively is very
important. Therefore, an Adaptive Data Augmentation Framework based on the
tensor T-product Operator is proposed in this paper, to triple one image data
to be trained and gain the result from all these three images together with
only less than 0.1% increase in the number of parameters. At the same time,
this framework serves the functions of column image embedding and global
feature intersection, enabling the model to obtain information in not only
spatial but frequency domain, and thus improving the prediction accuracy of the
model. Numerical experiments have been designed for several models, and the
results demonstrate the effectiveness of this adaptive framework. Numerical
experiments show that our data augmentation framework can improve the
performance of original neural network model by 2%, which provides competitive
results to state-of-the-art methods.
|
2306.04276 | J\"org Rychen | J\"org Rychen, Patrick Paitz, Pascal Edme, Krystyna Smolinski, Joeri
Brackenhoff, and Andreas Fichtner | Test experiments with distributed acoustic sensing and hydrophone arrays
for locating underwater sound sources | Data description | null | 10.5281/zenodo.7886409 | null | physics.ao-ph cs.SD eess.AS physics.bio-ph | http://creativecommons.org/licenses/by/4.0/ | Whales and dolphins rely on sound for navigation and communication, making
them an intriguing subject for studying language evolution. Traditional
hydrophone arrays have been used to record their acoustic behavior, but optical
fibers have emerged as a promising alternative. This study explores the use of
distributed acoustic sensing (DAS), a technique that detects local stress in
optical fibers, for underwater sound recording. An experiment was conducted in
Lake Zurich, where a fiber-optic cable and a self-made hydrophone array were
deployed. A test signal was broadcasted at various locations, and the resulting
data was synchronized and consolidated into files. Analysis revealed distinct
frequency responses in the DAS channels and provided insights into sound
propagation in the lake. Challenges related to cable sensitivity, sample rate,
and broadcast fidelity were identified. This dataset serves as a valuable
resource for advancing acoustic sensing techniques in underwater environments,
especially for studying marine mammal vocal behavior.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 09:22:28 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Rychen",
"Jörg",
""
],
[
"Paitz",
"Patrick",
""
],
[
"Edme",
"Pascal",
""
],
[
"Smolinski",
"Krystyna",
""
],
[
"Brackenhoff",
"Joeri",
""
],
[
"Fichtner",
"Andreas",
""
]
] | TITLE: Test experiments with distributed acoustic sensing and hydrophone arrays
for locating underwater sound sources
ABSTRACT: Whales and dolphins rely on sound for navigation and communication, making
them an intriguing subject for studying language evolution. Traditional
hydrophone arrays have been used to record their acoustic behavior, but optical
fibers have emerged as a promising alternative. This study explores the use of
distributed acoustic sensing (DAS), a technique that detects local stress in
optical fibers, for underwater sound recording. An experiment was conducted in
Lake Zurich, where a fiber-optic cable and a self-made hydrophone array were
deployed. A test signal was broadcasted at various locations, and the resulting
data was synchronized and consolidated into files. Analysis revealed distinct
frequency responses in the DAS channels and provided insights into sound
propagation in the lake. Challenges related to cable sensitivity, sample rate,
and broadcast fidelity were identified. This dataset serves as a valuable
resource for advancing acoustic sensing techniques in underwater environments,
especially for studying marine mammal vocal behavior.
|
2306.04285 | Isaiah Hull | Jes\'us Fern\'andez-Villaverde and Isaiah Hull | Dynamic Programming on a Quantum Annealer: Solving the RBC Model | 46 pages, 11 figures | null | null | null | econ.GN cs.CE q-fin.EC quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce a novel approach to solving dynamic programming problems, such
as those in many economic models, on a quantum annealer, a specialized device
that performs combinatorial optimization. Quantum annealers attempt to solve an
NP-hard problem by starting in a quantum superposition of all states and
generating candidate global solutions in milliseconds, irrespective of problem
size. Using existing quantum hardware, we achieve an order-of-magnitude
speed-up in solving the real business cycle model over benchmarks in the
literature. We also provide a detailed introduction to quantum annealing and
discuss its potential use for more challenging economic problems.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 09:38:42 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Fernández-Villaverde",
"Jesús",
""
],
[
"Hull",
"Isaiah",
""
]
] | TITLE: Dynamic Programming on a Quantum Annealer: Solving the RBC Model
ABSTRACT: We introduce a novel approach to solving dynamic programming problems, such
as those in many economic models, on a quantum annealer, a specialized device
that performs combinatorial optimization. Quantum annealers attempt to solve an
NP-hard problem by starting in a quantum superposition of all states and
generating candidate global solutions in milliseconds, irrespective of problem
size. Using existing quantum hardware, we achieve an order-of-magnitude
speed-up in solving the real business cycle model over benchmarks in the
literature. We also provide a detailed introduction to quantum annealing and
discuss its potential use for more challenging economic problems.
|
2306.04292 | Timo Freiesleben | Timo Freiesleben and Gunnar K\"onig | Dear XAI Community, We Need to Talk! Fundamental Misconceptions in
Current XAI Research | A revised version of this preprint has been accepted at the World XAI
Conference. It will be referenced as soon as it is published | null | null | null | cs.AI stat.ML | http://creativecommons.org/licenses/by/4.0/ | Despite progress in the field, significant parts of current XAI research are
still not on solid conceptual, ethical, or methodological grounds.
Unfortunately, these unfounded parts are not on the decline but continue to
grow. Many explanation techniques are still proposed without clarifying their
purpose. Instead, they are advertised with ever more fancy-looking heatmaps or
only seemingly relevant benchmarks. Moreover, explanation techniques are
motivated with questionable goals, such as building trust, or rely on strong
assumptions about the 'concepts' that deep learning algorithms learn. In this
paper, we highlight and discuss these and other misconceptions in current XAI
research. We also suggest steps to make XAI a more substantive area of
research.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 09:46:38 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Freiesleben",
"Timo",
""
],
[
"König",
"Gunnar",
""
]
] | TITLE: Dear XAI Community, We Need to Talk! Fundamental Misconceptions in
Current XAI Research
ABSTRACT: Despite progress in the field, significant parts of current XAI research are
still not on solid conceptual, ethical, or methodological grounds.
Unfortunately, these unfounded parts are not on the decline but continue to
grow. Many explanation techniques are still proposed without clarifying their
purpose. Instead, they are advertised with ever more fancy-looking heatmaps or
only seemingly relevant benchmarks. Moreover, explanation techniques are
motivated with questionable goals, such as building trust, or rely on strong
assumptions about the 'concepts' that deep learning algorithms learn. In this
paper, we highlight and discuss these and other misconceptions in current XAI
research. We also suggest steps to make XAI a more substantive area of
research.
|
2306.04293 | Soyeong Jeong | Soyeong Jeong, Jinheon Baek, Sung Ju Hwang, Jong C. Park | Phrase Retrieval for Open-Domain Conversational Question Answering with
Conversational Dependency Modeling via Contrastive Learning | Findings of ACL 2023 | null | null | null | cs.CL cs.IR cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Open-Domain Conversational Question Answering (ODConvQA) aims at answering
questions through a multi-turn conversation based on a retriever-reader
pipeline, which retrieves passages and then predicts answers with them.
However, such a pipeline approach not only makes the reader vulnerable to the
errors propagated from the retriever, but also demands additional effort to
develop both the retriever and the reader, which further makes it slower since
they are not runnable in parallel. In this work, we propose a method to
directly predict answers with a phrase retrieval scheme for a sequence of
words, reducing the conventional two distinct subtasks into a single one. Also,
for the first time, we study its capability for ODConvQA tasks. However, simply
adopting it is largely problematic, due to the dependencies between previous
and current turns in a conversation. To address this problem, we further
introduce a novel contrastive learning strategy, making sure to reflect
previous turns when retrieving the phrase for the current context, by
maximizing representational similarities of consecutive turns in a conversation
while minimizing irrelevant conversational contexts. We validate our model on
two ODConvQA datasets, whose experimental results show that it substantially
outperforms the relevant baselines with the retriever-reader. Code is available
at: https://github.com/starsuzi/PRO-ConvQA.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 09:46:38 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Jeong",
"Soyeong",
""
],
[
"Baek",
"Jinheon",
""
],
[
"Hwang",
"Sung Ju",
""
],
[
"Park",
"Jong C.",
""
]
] | TITLE: Phrase Retrieval for Open-Domain Conversational Question Answering with
Conversational Dependency Modeling via Contrastive Learning
ABSTRACT: Open-Domain Conversational Question Answering (ODConvQA) aims at answering
questions through a multi-turn conversation based on a retriever-reader
pipeline, which retrieves passages and then predicts answers with them.
However, such a pipeline approach not only makes the reader vulnerable to the
errors propagated from the retriever, but also demands additional effort to
develop both the retriever and the reader, which further makes it slower since
they are not runnable in parallel. In this work, we propose a method to
directly predict answers with a phrase retrieval scheme for a sequence of
words, reducing the conventional two distinct subtasks into a single one. Also,
for the first time, we study its capability for ODConvQA tasks. However, simply
adopting it is largely problematic, due to the dependencies between previous
and current turns in a conversation. To address this problem, we further
introduce a novel contrastive learning strategy, making sure to reflect
previous turns when retrieving the phrase for the current context, by
maximizing representational similarities of consecutive turns in a conversation
while minimizing irrelevant conversational contexts. We validate our model on
two ODConvQA datasets, whose experimental results show that it substantially
outperforms the relevant baselines with the retriever-reader. Code is available
at: https://github.com/starsuzi/PRO-ConvQA.
|
2306.04314 | Gil Rocha | Gil Rocha, Henrique Lopes Cardoso, Jonas Belouadi, Steffen Eger | Cross-Genre Argument Mining: Can Language Models Automatically Fill in
Missing Discourse Markers? | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Available corpora for Argument Mining differ along several axes, and one of
the key differences is the presence (or absence) of discourse markers to signal
argumentative content. Exploring effective ways to use discourse markers has
received wide attention in various discourse parsing tasks, from which it is
well-known that discourse markers are strong indicators of discourse relations.
To improve the robustness of Argument Mining systems across different genres,
we propose to automatically augment a given text with discourse markers such
that all relations are explicitly signaled. Our analysis unveils that popular
language models taken out-of-the-box fail on this task; however, when
fine-tuned on a new heterogeneous dataset that we construct (including
synthetic and real examples), they perform considerably better. We demonstrate
the impact of our approach on an Argument Mining downstream task, evaluated on
different corpora, showing that language models can be trained to automatically
fill in discourse markers across different corpora, improving the performance
of a downstream model in some, but not all, cases. Our proposed approach can
further be employed as an assistive tool for better discourse understanding.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 10:19:50 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Rocha",
"Gil",
""
],
[
"Cardoso",
"Henrique Lopes",
""
],
[
"Belouadi",
"Jonas",
""
],
[
"Eger",
"Steffen",
""
]
] | TITLE: Cross-Genre Argument Mining: Can Language Models Automatically Fill in
Missing Discourse Markers?
ABSTRACT: Available corpora for Argument Mining differ along several axes, and one of
the key differences is the presence (or absence) of discourse markers to signal
argumentative content. Exploring effective ways to use discourse markers has
received wide attention in various discourse parsing tasks, from which it is
well-known that discourse markers are strong indicators of discourse relations.
To improve the robustness of Argument Mining systems across different genres,
we propose to automatically augment a given text with discourse markers such
that all relations are explicitly signaled. Our analysis unveils that popular
language models taken out-of-the-box fail on this task; however, when
fine-tuned on a new heterogeneous dataset that we construct (including
synthetic and real examples), they perform considerably better. We demonstrate
the impact of our approach on an Argument Mining downstream task, evaluated on
different corpora, showing that language models can be trained to automatically
fill in discourse markers across different corpora, improving the performance
of a downstream model in some, but not all, cases. Our proposed approach can
further be employed as an assistive tool for better discourse understanding.
|
2306.04333 | Wenyan Ma | Wenyan Ma, Lipeng Zhu, Rui Zhang | Compressed Sensing Based Channel Estimation for Movable Antenna
Communications | null | null | null | null | cs.IT eess.SP math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this letter, we study the channel estimation for wireless communications
with movable antenna (MA), which requires to reconstruct the channel response
at any location in a given region where the transmitter/receiver is located
based on the channel measurements taken at finite locations therein, so as to
find the MA's location for optimizing the communication performance. To reduce
the pilot overhead and computational complexity for channel estimation, we
propose a new successive transmitter-receiver compressed sensing (STRCS) method
by exploiting the efficient representation of the channel responses in the
given transmitter/receiver region (field) in terms of multi-path components.
Specifically, the field-response information (FRI) in the angular domain,
including the angles of departure (AoDs)/angles of arrival (AoAs) and complex
coefficients of all significant multi-path components are sequentially
estimated based on a finite number of channel measurements taken at
random/selected locations by the MA at the transmitter and/or receiver.
Simulation results demonstrate that the proposed channel reconstruction method
outperforms the benchmark schemes in terms of both pilot overhead and channel
reconstruction accuracy.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 10:59:34 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Ma",
"Wenyan",
""
],
[
"Zhu",
"Lipeng",
""
],
[
"Zhang",
"Rui",
""
]
] | TITLE: Compressed Sensing Based Channel Estimation for Movable Antenna
Communications
ABSTRACT: In this letter, we study the channel estimation for wireless communications
with movable antenna (MA), which requires to reconstruct the channel response
at any location in a given region where the transmitter/receiver is located
based on the channel measurements taken at finite locations therein, so as to
find the MA's location for optimizing the communication performance. To reduce
the pilot overhead and computational complexity for channel estimation, we
propose a new successive transmitter-receiver compressed sensing (STRCS) method
by exploiting the efficient representation of the channel responses in the
given transmitter/receiver region (field) in terms of multi-path components.
Specifically, the field-response information (FRI) in the angular domain,
including the angles of departure (AoDs)/angles of arrival (AoAs) and complex
coefficients of all significant multi-path components are sequentially
estimated based on a finite number of channel measurements taken at
random/selected locations by the MA at the transmitter and/or receiver.
Simulation results demonstrate that the proposed channel reconstruction method
outperforms the benchmark schemes in terms of both pilot overhead and channel
reconstruction accuracy.
|
2306.04334 | Alessandro Scir\`e | Alessandro Scir\`e, Simone Conia, Simone Ciciliano, Roberto Navigli | Echoes from Alexandria: A Large Resource for Multilingual Book
Summarization | 9 pages, long paper at ACL 2023 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | In recent years, research in text summarization has mainly focused on the
news domain, where texts are typically short and have strong layout features.
The task of full-book summarization presents additional challenges which are
hard to tackle with current resources, due to their limited size and
availability in English only. To overcome these limitations, we present "Echoes
from Alexandria", or in shortened form, "Echoes", a large resource for
multilingual book summarization. Echoes features three novel datasets: i)
Echo-Wiki, for multilingual book summarization, ii) Echo-XSum, for
extremely-compressive multilingual book summarization, and iii) Echo-FairySum,
for extractive book summarization. To the best of our knowledge, Echoes, with
its thousands of books and summaries, is the largest resource, and the first to
be multilingual, featuring 5 languages and 25 language pairs. In addition to
Echoes, we also introduce a new extractive-then-abstractive baseline, and,
supported by our experimental results and manual analysis of the summaries
generated, we argue that this baseline is more suitable for book summarization
than purely-abstractive approaches. We release our resource and software at
https://github.com/Babelscape/echoes-from-alexandria in the hope of fostering
innovative research in multilingual book summarization.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 11:01:39 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Scirè",
"Alessandro",
""
],
[
"Conia",
"Simone",
""
],
[
"Ciciliano",
"Simone",
""
],
[
"Navigli",
"Roberto",
""
]
] | TITLE: Echoes from Alexandria: A Large Resource for Multilingual Book
Summarization
ABSTRACT: In recent years, research in text summarization has mainly focused on the
news domain, where texts are typically short and have strong layout features.
The task of full-book summarization presents additional challenges which are
hard to tackle with current resources, due to their limited size and
availability in English only. To overcome these limitations, we present "Echoes
from Alexandria", or in shortened form, "Echoes", a large resource for
multilingual book summarization. Echoes features three novel datasets: i)
Echo-Wiki, for multilingual book summarization, ii) Echo-XSum, for
extremely-compressive multilingual book summarization, and iii) Echo-FairySum,
for extractive book summarization. To the best of our knowledge, Echoes, with
its thousands of books and summaries, is the largest resource, and the first to
be multilingual, featuring 5 languages and 25 language pairs. In addition to
Echoes, we also introduce a new extractive-then-abstractive baseline, and,
supported by our experimental results and manual analysis of the summaries
generated, we argue that this baseline is more suitable for book summarization
than purely-abstractive approaches. We release our resource and software at
https://github.com/Babelscape/echoes-from-alexandria in the hope of fostering
innovative research in multilingual book summarization.
|
2306.04353 | Nicola Asuni | Nicola Asuni | Reversible Numeric Composite Key (RNCK) | 5 pages | null | null | null | cs.DB | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In database design, Composite Keys are used to uniquely identify records and
prevent data duplication. However, they require more memory and storage space
than single keys, and can make queries more CPU-intensive. Surrogate Keys are
an alternative that can overcome some of these limitations, but they can also
introduce new disadvantages. To address these challenges, a new type of key
called a Reversible Numeric Composite Key (RNCK) has been developed. RNCK is a
single number that encodes multiple data attributes, and can be decoded back to
the original values. This makes it possible to achieve the benefits of both
Composite Keys and Surrogate Keys, while overcoming some of their limitations.
RNCK has been shown to improve query performance and reduce memory and storage
requirements. It can be used in relational databases, large static datasets,
and key-value caching systems. RNCK has been successfully used in production
systems for several years.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 11:35:46 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Asuni",
"Nicola",
""
]
] | TITLE: Reversible Numeric Composite Key (RNCK)
ABSTRACT: In database design, Composite Keys are used to uniquely identify records and
prevent data duplication. However, they require more memory and storage space
than single keys, and can make queries more CPU-intensive. Surrogate Keys are
an alternative that can overcome some of these limitations, but they can also
introduce new disadvantages. To address these challenges, a new type of key
called a Reversible Numeric Composite Key (RNCK) has been developed. RNCK is a
single number that encodes multiple data attributes, and can be decoded back to
the original values. This makes it possible to achieve the benefits of both
Composite Keys and Surrogate Keys, while overcoming some of their limitations.
RNCK has been shown to improve query performance and reduce memory and storage
requirements. It can be used in relational databases, large static datasets,
and key-value caching systems. RNCK has been successfully used in production
systems for several years.
|
2306.04362 | Qinghao Ye | Haiyang Xu, Qinghao Ye, Xuan Wu, Ming Yan, Yuan Miao, Jiabo Ye, Guohai
Xu, Anwen Hu, Yaya Shi, Guangwei Xu, Chenliang Li, Qi Qian, Maofei Que, Ji
Zhang, Xiao Zeng, Fei Huang | Youku-mPLUG: A 10 Million Large-scale Chinese Video-Language Dataset for
Pre-training and Benchmarks | Working in progress | null | null | null | cs.CV cs.CL | http://creativecommons.org/licenses/by-nc-sa/4.0/ | To promote the development of Vision-Language Pre-training (VLP) and
multimodal Large Language Model (LLM) in the Chinese community, we firstly
release the largest public Chinese high-quality video-language dataset named
Youku-mPLUG, which is collected from Youku, a well-known Chinese video-sharing
website, with strict criteria of safety, diversity, and quality. Youku-mPLUG
contains 10 million Chinese video-text pairs filtered from 400 million raw
videos across a wide range of 45 diverse categories for large-scale
pre-training. In addition, to facilitate a comprehensive evaluation of
video-language models, we carefully build the largest human-annotated Chinese
benchmarks covering three popular video-language tasks of cross-modal
retrieval, video captioning, and video category classification. Youku-mPLUG can
enable researchers to conduct more in-depth multimodal research and develop
better applications in the future. Furthermore, we release popular
video-language pre-training models, ALPRO and mPLUG-2, and our proposed
modularized decoder-only model mPLUG-video pre-trained on Youku-mPLUG.
Experiments show that models pre-trained on Youku-mPLUG gain up to 23.1%
improvement in video category classification. Besides, mPLUG-video achieves a
new state-of-the-art result on these benchmarks with 80.5% top-1 accuracy in
video category classification and 68.9 CIDEr score in video captioning,
respectively. Finally, we scale up mPLUG-video based on the frozen Bloomz with
only 1.7% trainable parameters as Chinese multimodal LLM, and demonstrate
impressive instruction and video understanding ability. The zero-shot
instruction understanding experiment indicates that pretraining with
Youku-mPLUG can enhance the ability to comprehend overall and detailed visual
semantics, recognize scene text, and leverage open-domain knowledge.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 11:52:36 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Xu",
"Haiyang",
""
],
[
"Ye",
"Qinghao",
""
],
[
"Wu",
"Xuan",
""
],
[
"Yan",
"Ming",
""
],
[
"Miao",
"Yuan",
""
],
[
"Ye",
"Jiabo",
""
],
[
"Xu",
"Guohai",
""
],
[
"Hu",
"Anwen",
""
],
[
"Shi",
"Yaya",
""
],
[
"Xu",
"Guangwei",
""
],
[
"Li",
"Chenliang",
""
],
[
"Qian",
"Qi",
""
],
[
"Que",
"Maofei",
""
],
[
"Zhang",
"Ji",
""
],
[
"Zeng",
"Xiao",
""
],
[
"Huang",
"Fei",
""
]
] | TITLE: Youku-mPLUG: A 10 Million Large-scale Chinese Video-Language Dataset for
Pre-training and Benchmarks
ABSTRACT: To promote the development of Vision-Language Pre-training (VLP) and
multimodal Large Language Model (LLM) in the Chinese community, we firstly
release the largest public Chinese high-quality video-language dataset named
Youku-mPLUG, which is collected from Youku, a well-known Chinese video-sharing
website, with strict criteria of safety, diversity, and quality. Youku-mPLUG
contains 10 million Chinese video-text pairs filtered from 400 million raw
videos across a wide range of 45 diverse categories for large-scale
pre-training. In addition, to facilitate a comprehensive evaluation of
video-language models, we carefully build the largest human-annotated Chinese
benchmarks covering three popular video-language tasks of cross-modal
retrieval, video captioning, and video category classification. Youku-mPLUG can
enable researchers to conduct more in-depth multimodal research and develop
better applications in the future. Furthermore, we release popular
video-language pre-training models, ALPRO and mPLUG-2, and our proposed
modularized decoder-only model mPLUG-video pre-trained on Youku-mPLUG.
Experiments show that models pre-trained on Youku-mPLUG gain up to 23.1%
improvement in video category classification. Besides, mPLUG-video achieves a
new state-of-the-art result on these benchmarks with 80.5% top-1 accuracy in
video category classification and 68.9 CIDEr score in video captioning,
respectively. Finally, we scale up mPLUG-video based on the frozen Bloomz with
only 1.7% trainable parameters as Chinese multimodal LLM, and demonstrate
impressive instruction and video understanding ability. The zero-shot
instruction understanding experiment indicates that pretraining with
Youku-mPLUG can enhance the ability to comprehend overall and detailed visual
semantics, recognize scene text, and leverage open-domain knowledge.
|
2306.04368 | Massa Baali | Massa Baali, Ibrahim Almakky, Shady Shehata, Fakhri Karray | Arabic Dysarthric Speech Recognition Using Adversarial and Signal-Based
Augmentation | Accepted to Interspeech 2023 | null | null | null | cs.SD cs.CL eess.AS | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Despite major advancements in Automatic Speech Recognition (ASR), the
state-of-the-art ASR systems struggle to deal with impaired speech even with
high-resource languages. In Arabic, this challenge gets amplified, with added
complexities in collecting data from dysarthric speakers. In this paper, we aim
to improve the performance of Arabic dysarthric automatic speech recognition
through a multi-stage augmentation approach. To this effect, we first propose a
signal-based approach to generate dysarthric Arabic speech from healthy Arabic
speech by modifying its speed and tempo. We also propose a second stage
Parallel Wave Generative (PWG) adversarial model that is trained on an English
dysarthric dataset to capture language-independant dysarthric speech patterns
and further augment the signal-adjusted speech samples. Furthermore, we propose
a fine-tuning and text-correction strategies for Arabic Conformer at different
dysarthric speech severity levels. Our fine-tuned Conformer achieved 18% Word
Error Rate (WER) and 17.2% Character Error Rate (CER) on synthetically
generated dysarthric speech from the Arabic commonvoice speech dataset. This
shows significant WER improvement of 81.8% compared to the baseline model
trained solely on healthy data. We perform further validation on real English
dysarthric speech showing a WER improvement of 124% compared to the baseline
trained only on healthy English LJSpeech dataset.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 12:01:46 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Baali",
"Massa",
""
],
[
"Almakky",
"Ibrahim",
""
],
[
"Shehata",
"Shady",
""
],
[
"Karray",
"Fakhri",
""
]
] | TITLE: Arabic Dysarthric Speech Recognition Using Adversarial and Signal-Based
Augmentation
ABSTRACT: Despite major advancements in Automatic Speech Recognition (ASR), the
state-of-the-art ASR systems struggle to deal with impaired speech even with
high-resource languages. In Arabic, this challenge gets amplified, with added
complexities in collecting data from dysarthric speakers. In this paper, we aim
to improve the performance of Arabic dysarthric automatic speech recognition
through a multi-stage augmentation approach. To this effect, we first propose a
signal-based approach to generate dysarthric Arabic speech from healthy Arabic
speech by modifying its speed and tempo. We also propose a second stage
Parallel Wave Generative (PWG) adversarial model that is trained on an English
dysarthric dataset to capture language-independant dysarthric speech patterns
and further augment the signal-adjusted speech samples. Furthermore, we propose
a fine-tuning and text-correction strategies for Arabic Conformer at different
dysarthric speech severity levels. Our fine-tuned Conformer achieved 18% Word
Error Rate (WER) and 17.2% Character Error Rate (CER) on synthetically
generated dysarthric speech from the Arabic commonvoice speech dataset. This
shows significant WER improvement of 81.8% compared to the baseline model
trained solely on healthy data. We perform further validation on real English
dysarthric speech showing a WER improvement of 124% compared to the baseline
trained only on healthy English LJSpeech dataset.
|
2306.04374 | Shikhar Vashishth | Shikhar Vashishth, Shikhar Bharadwaj, Sriram Ganapathy, Ankur Bapna,
Min Ma, Wei Han, Vera Axelrod, Partha Talukdar | Label Aware Speech Representation Learning For Language Identification | Accepted at Interspeech 2023 | null | null | null | cs.CL cs.LG cs.SD eess.AS | http://creativecommons.org/licenses/by/4.0/ | Speech representation learning approaches for non-semantic tasks such as
language recognition have either explored supervised embedding extraction
methods using a classifier model or self-supervised representation learning
approaches using raw data. In this paper, we propose a novel framework of
combining self-supervised representation learning with the language label
information for the pre-training task. This framework, termed as Label Aware
Speech Representation (LASR) learning, uses a triplet based objective function
to incorporate language labels along with the self-supervised loss function.
The speech representations are further fine-tuned for the downstream task. The
language recognition experiments are performed on two public datasets - FLEURS
and Dhwani. In these experiments, we illustrate that the proposed LASR
framework improves over the state-of-the-art systems on language
identification. We also report an analysis of the robustness of LASR approach
to noisy/missing labels as well as its application to multi-lingual speech
recognition tasks.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 12:14:16 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Vashishth",
"Shikhar",
""
],
[
"Bharadwaj",
"Shikhar",
""
],
[
"Ganapathy",
"Sriram",
""
],
[
"Bapna",
"Ankur",
""
],
[
"Ma",
"Min",
""
],
[
"Han",
"Wei",
""
],
[
"Axelrod",
"Vera",
""
],
[
"Talukdar",
"Partha",
""
]
] | TITLE: Label Aware Speech Representation Learning For Language Identification
ABSTRACT: Speech representation learning approaches for non-semantic tasks such as
language recognition have either explored supervised embedding extraction
methods using a classifier model or self-supervised representation learning
approaches using raw data. In this paper, we propose a novel framework of
combining self-supervised representation learning with the language label
information for the pre-training task. This framework, termed as Label Aware
Speech Representation (LASR) learning, uses a triplet based objective function
to incorporate language labels along with the self-supervised loss function.
The speech representations are further fine-tuned for the downstream task. The
language recognition experiments are performed on two public datasets - FLEURS
and Dhwani. In these experiments, we illustrate that the proposed LASR
framework improves over the state-of-the-art systems on language
identification. We also report an analysis of the robustness of LASR approach
to noisy/missing labels as well as its application to multi-lingual speech
recognition tasks.
|
2306.04384 | Xavier Fontaine | Xavier Fontaine, F\'elix Gaschi, Parisa Rastin and Yannick Toussaint | Multilingual Clinical NER: Translation or Cross-lingual Transfer? | 23 pages, Proceedings of the 5th Clinical Natural Language Processing
Workshop | null | null | null | cs.CL cs.AI cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Natural language tasks like Named Entity Recognition (NER) in the clinical
domain on non-English texts can be very time-consuming and expensive due to the
lack of annotated data. Cross-lingual transfer (CLT) is a way to circumvent
this issue thanks to the ability of multilingual large language models to be
fine-tuned on a specific task in one language and to provide high accuracy for
the same task in another language. However, other methods leveraging
translation models can be used to perform NER without annotated data in the
target language, by either translating the training set or test set. This paper
compares cross-lingual transfer with these two alternative methods, to perform
clinical NER in French and in German without any training data in those
languages. To this end, we release MedNERF a medical NER test set extracted
from French drug prescriptions and annotated with the same guidelines as an
English dataset. Through extensive experiments on this dataset and on a German
medical dataset (Frei and Kramer, 2021), we show that translation-based methods
can achieve similar performance to CLT but require more care in their design.
And while they can take advantage of monolingual clinical language models,
those do not guarantee better results than large general-purpose multilingual
models, whether with cross-lingual transfer or translation.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 12:31:07 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Fontaine",
"Xavier",
""
],
[
"Gaschi",
"Félix",
""
],
[
"Rastin",
"Parisa",
""
],
[
"Toussaint",
"Yannick",
""
]
] | TITLE: Multilingual Clinical NER: Translation or Cross-lingual Transfer?
ABSTRACT: Natural language tasks like Named Entity Recognition (NER) in the clinical
domain on non-English texts can be very time-consuming and expensive due to the
lack of annotated data. Cross-lingual transfer (CLT) is a way to circumvent
this issue thanks to the ability of multilingual large language models to be
fine-tuned on a specific task in one language and to provide high accuracy for
the same task in another language. However, other methods leveraging
translation models can be used to perform NER without annotated data in the
target language, by either translating the training set or test set. This paper
compares cross-lingual transfer with these two alternative methods, to perform
clinical NER in French and in German without any training data in those
languages. To this end, we release MedNERF a medical NER test set extracted
from French drug prescriptions and annotated with the same guidelines as an
English dataset. Through extensive experiments on this dataset and on a German
medical dataset (Frei and Kramer, 2021), we show that translation-based methods
can achieve similar performance to CLT but require more care in their design.
And while they can take advantage of monolingual clinical language models,
those do not guarantee better results than large general-purpose multilingual
models, whether with cross-lingual transfer or translation.
|
2306.04385 | Han Sun | Han Sun, Rui Gong, Konrad Schindler, Luc Van Gool | SF-FSDA: Source-Free Few-Shot Domain Adaptive Object Detection with
Efficient Labeled Data Factory | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Domain adaptive object detection aims to leverage the knowledge learned from
a labeled source domain to improve the performance on an unlabeled target
domain. Prior works typically require the access to the source domain data for
adaptation, and the availability of sufficient data on the target domain.
However, these assumptions may not hold due to data privacy and rare data
collection. In this paper, we propose and investigate a more practical and
challenging domain adaptive object detection problem under both source-free and
few-shot conditions, named as SF-FSDA. To overcome this problem, we develop an
efficient labeled data factory based approach. Without accessing the source
domain, the data factory renders i) infinite amount of synthesized
target-domain like images, under the guidance of the few-shot image samples and
text description from the target domain; ii) corresponding bounding box and
category annotations, only demanding minimum human effort, i.e., a few manually
labeled examples. On the one hand, the synthesized images mitigate the
knowledge insufficiency brought by the few-shot condition. On the other hand,
compared to the popular pseudo-label technique, the generated annotations from
data factory not only get rid of the reliance on the source pretrained object
detection model, but also alleviate the unavoidably pseudo-label noise due to
domain shift and source-free condition. The generated dataset is further
utilized to adapt the source pretrained object detection model, realizing the
robust object detection under SF-FSDA. The experiments on different settings
showcase that our proposed approach outperforms other state-of-the-art methods
on SF-FSDA problem. Our codes and models will be made publicly available.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 12:34:55 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Sun",
"Han",
""
],
[
"Gong",
"Rui",
""
],
[
"Schindler",
"Konrad",
""
],
[
"Van Gool",
"Luc",
""
]
] | TITLE: SF-FSDA: Source-Free Few-Shot Domain Adaptive Object Detection with
Efficient Labeled Data Factory
ABSTRACT: Domain adaptive object detection aims to leverage the knowledge learned from
a labeled source domain to improve the performance on an unlabeled target
domain. Prior works typically require the access to the source domain data for
adaptation, and the availability of sufficient data on the target domain.
However, these assumptions may not hold due to data privacy and rare data
collection. In this paper, we propose and investigate a more practical and
challenging domain adaptive object detection problem under both source-free and
few-shot conditions, named as SF-FSDA. To overcome this problem, we develop an
efficient labeled data factory based approach. Without accessing the source
domain, the data factory renders i) infinite amount of synthesized
target-domain like images, under the guidance of the few-shot image samples and
text description from the target domain; ii) corresponding bounding box and
category annotations, only demanding minimum human effort, i.e., a few manually
labeled examples. On the one hand, the synthesized images mitigate the
knowledge insufficiency brought by the few-shot condition. On the other hand,
compared to the popular pseudo-label technique, the generated annotations from
data factory not only get rid of the reliance on the source pretrained object
detection model, but also alleviate the unavoidably pseudo-label noise due to
domain shift and source-free condition. The generated dataset is further
utilized to adapt the source pretrained object detection model, realizing the
robust object detection under SF-FSDA. The experiments on different settings
showcase that our proposed approach outperforms other state-of-the-art methods
on SF-FSDA problem. Our codes and models will be made publicly available.
|
2306.04433 | Naimul Mefraz Khan | Md Niaz Imtiaz and Naimul Khan | Cross-Database and Cross-Channel ECG Arrhythmia Heartbeat Classification
Based on Unsupervised Domain Adaptation | null | null | null | null | eess.SP cs.AI | http://creativecommons.org/licenses/by/4.0/ | The classification of electrocardiogram (ECG) plays a crucial role in the
development of an automatic cardiovascular diagnostic system. However,
considerable variances in ECG signals between individuals is a significant
challenge. Changes in data distribution limit cross-domain utilization of a
model. In this study, we propose a solution to classify ECG in an unlabeled
dataset by leveraging knowledge obtained from labeled source domain. We present
a domain-adaptive deep network based on cross-domain feature discrepancy
optimization. Our method comprises three stages: pre-training, cluster-centroid
computing, and adaptation. In pre-training, we employ a Distributionally Robust
Optimization (DRO) technique to deal with the vanishing worst-case training
loss. To enhance the richness of the features, we concatenate three temporal
features with the deep learning features. The cluster computing stage involves
computing centroids of distinctly separable clusters for the source using true
labels, and for the target using confident predictions. We propose a novel
technique to select confident predictions in the target domain. In the
adaptation stage, we minimize compacting loss within the same cluster,
separating loss across different clusters, inter-domain cluster discrepancy
loss, and running combined loss to produce a domain-robust model. Experiments
conducted in both cross-domain and cross-channel paradigms show the efficacy of
the proposed method. Our method achieves superior performance compared to other
state-of-the-art approaches in detecting ventricular ectopic beats (V),
supraventricular ectopic beats (S), and fusion beats (F). Our method achieves
an average improvement of 11.78% in overall accuracy over the
non-domain-adaptive baseline method on the three test datasets.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 13:46:49 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Imtiaz",
"Md Niaz",
""
],
[
"Khan",
"Naimul",
""
]
] | TITLE: Cross-Database and Cross-Channel ECG Arrhythmia Heartbeat Classification
Based on Unsupervised Domain Adaptation
ABSTRACT: The classification of electrocardiogram (ECG) plays a crucial role in the
development of an automatic cardiovascular diagnostic system. However,
considerable variances in ECG signals between individuals is a significant
challenge. Changes in data distribution limit cross-domain utilization of a
model. In this study, we propose a solution to classify ECG in an unlabeled
dataset by leveraging knowledge obtained from labeled source domain. We present
a domain-adaptive deep network based on cross-domain feature discrepancy
optimization. Our method comprises three stages: pre-training, cluster-centroid
computing, and adaptation. In pre-training, we employ a Distributionally Robust
Optimization (DRO) technique to deal with the vanishing worst-case training
loss. To enhance the richness of the features, we concatenate three temporal
features with the deep learning features. The cluster computing stage involves
computing centroids of distinctly separable clusters for the source using true
labels, and for the target using confident predictions. We propose a novel
technique to select confident predictions in the target domain. In the
adaptation stage, we minimize compacting loss within the same cluster,
separating loss across different clusters, inter-domain cluster discrepancy
loss, and running combined loss to produce a domain-robust model. Experiments
conducted in both cross-domain and cross-channel paradigms show the efficacy of
the proposed method. Our method achieves superior performance compared to other
state-of-the-art approaches in detecting ventricular ectopic beats (V),
supraventricular ectopic beats (S), and fusion beats (F). Our method achieves
an average improvement of 11.78% in overall accuracy over the
non-domain-adaptive baseline method on the three test datasets.
|
2306.04441 | Weizhi Wang | Weizhi Wang, Hong Wang, Xifeng Yan | STEPS: A Benchmark for Order Reasoning in Sequential Tasks | Work in Progress | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Various human activities can be abstracted into a sequence of actions in
natural text, i.e. cooking, repairing, manufacturing, etc. Such action
sequences heavily depend on the executing order, while disorder in action
sequences leads to failure of further task execution by robots or AI agents.
Therefore, to verify the order reasoning capability of current neural models in
sequential tasks, we propose a challenging benchmark , named STEPS. STEPS
involves two subtask settings, focusing on determining the rationality of given
next step in recipes and selecting the reasonable step from the multi-choice
question, respectively. We describe the data construction and task
formulations, and benchmark most of significant Large Language Models (LLMs).
The experimental results demonstrate 1) The commonsense reasoning of action
orders in sequential tasks are challenging to resolve via zero-shot prompting
or few-shot in-context learning for LLMs; 2) Prompting method still
significantly lags behind tuning-based method on STEPS.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 13:58:55 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Wang",
"Weizhi",
""
],
[
"Wang",
"Hong",
""
],
[
"Yan",
"Xifeng",
""
]
] | TITLE: STEPS: A Benchmark for Order Reasoning in Sequential Tasks
ABSTRACT: Various human activities can be abstracted into a sequence of actions in
natural text, i.e. cooking, repairing, manufacturing, etc. Such action
sequences heavily depend on the executing order, while disorder in action
sequences leads to failure of further task execution by robots or AI agents.
Therefore, to verify the order reasoning capability of current neural models in
sequential tasks, we propose a challenging benchmark , named STEPS. STEPS
involves two subtask settings, focusing on determining the rationality of given
next step in recipes and selecting the reasonable step from the multi-choice
question, respectively. We describe the data construction and task
formulations, and benchmark most of significant Large Language Models (LLMs).
The experimental results demonstrate 1) The commonsense reasoning of action
orders in sequential tasks are challenging to resolve via zero-shot prompting
or few-shot in-context learning for LLMs; 2) Prompting method still
significantly lags behind tuning-based method on STEPS.
|
2306.04466 | Tengjiao He | Tengjiao He and Wenguang Wang | Point Cloud Video Anomaly Detection Based on Point Spatio-Temporal
Auto-Encoder | null | null | null | null | cs.CV eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Video anomaly detection has great potential in enhancing safety in the
production and monitoring of crucial areas. Currently, most video anomaly
detection methods are based on RGB modality, but its redundant semantic
information may breach the privacy of residents or patients. The 3D data
obtained by depth camera and LiDAR can accurately locate anomalous events in 3D
space while preserving human posture and motion information. Identifying
individuals through the point cloud is difficult due to its sparsity, which
protects personal privacy. In this study, we propose Point Spatio-Temporal
Auto-Encoder (PSTAE), an autoencoder framework that uses point cloud videos as
input to detect anomalies in point cloud videos. We introduce PSTOp and
PSTTransOp to maintain spatial geometric and temporal motion information in
point cloud videos. To measure the reconstruction loss of the proposed
autoencoder framework, we propose a reconstruction loss measurement strategy
based on a shallow feature extractor. Experimental results on the TIMo dataset
show that our method outperforms currently representative depth modality-based
methods in terms of AUROC and has superior performance in detecting Medical
Issue anomalies. These results suggest the potential of point cloud modality in
video anomaly detection. Our method sets a new state-of-the-art (SOTA) on the
TIMo dataset.
| [
{
"version": "v1",
"created": "Sun, 4 Jun 2023 10:30:28 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"He",
"Tengjiao",
""
],
[
"Wang",
"Wenguang",
""
]
] | TITLE: Point Cloud Video Anomaly Detection Based on Point Spatio-Temporal
Auto-Encoder
ABSTRACT: Video anomaly detection has great potential in enhancing safety in the
production and monitoring of crucial areas. Currently, most video anomaly
detection methods are based on RGB modality, but its redundant semantic
information may breach the privacy of residents or patients. The 3D data
obtained by depth camera and LiDAR can accurately locate anomalous events in 3D
space while preserving human posture and motion information. Identifying
individuals through the point cloud is difficult due to its sparsity, which
protects personal privacy. In this study, we propose Point Spatio-Temporal
Auto-Encoder (PSTAE), an autoencoder framework that uses point cloud videos as
input to detect anomalies in point cloud videos. We introduce PSTOp and
PSTTransOp to maintain spatial geometric and temporal motion information in
point cloud videos. To measure the reconstruction loss of the proposed
autoencoder framework, we propose a reconstruction loss measurement strategy
based on a shallow feature extractor. Experimental results on the TIMo dataset
show that our method outperforms currently representative depth modality-based
methods in terms of AUROC and has superior performance in detecting Medical
Issue anomalies. These results suggest the potential of point cloud modality in
video anomaly detection. Our method sets a new state-of-the-art (SOTA) on the
TIMo dataset.
|
2306.04474 | Lujian Yao | Lujian Yao, Haitao Zhao, Jingchao Peng, Zhongze Wang, Kaijie Zhao | FoSp: Focus and Separation Network for Early Smoke Segmentation | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Early smoke segmentation (ESS) enables the accurate identification of smoke
sources, facilitating the prompt extinguishing of fires and preventing
large-scale gas leaks. But ESS poses greater challenges than conventional
object and regular smoke segmentation due to its small scale and transparent
appearance, which can result in high miss detection rate and low precision. To
address these issues, a Focus and Separation Network (FoSp) is proposed. We
first introduce a Focus module employing bidirectional cascade which guides
low-resolution and high-resolution features towards mid-resolution to locate
and determine the scope of smoke, reducing the miss detection rate. Next, we
propose a Separation module that separates smoke images into a pure smoke
foreground and a smoke-free background, enhancing the contrast between smoke
and background fundamentally, improving segmentation precision. Finally, a
Domain Fusion module is developed to integrate the distinctive features of the
two modules which can balance recall and precision to achieve high F_beta.
Futhermore, to promote the development of ESS, we introduce a high-quality
real-world dataset called SmokeSeg, which contains more small and transparent
smoke than the existing datasets. Experimental results show that our model
achieves the best performance on three available datasets: SYN70K (mIoU:
83.00%), SMOKE5K (F_beta: 81.6%) and SmokeSeg (F_beta: 72.05%). Especially, our
FoSp outperforms SegFormer by 7.71% (F_beta) for early smoke segmentation on
SmokeSeg.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 14:45:24 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Yao",
"Lujian",
""
],
[
"Zhao",
"Haitao",
""
],
[
"Peng",
"Jingchao",
""
],
[
"Wang",
"Zhongze",
""
],
[
"Zhao",
"Kaijie",
""
]
] | TITLE: FoSp: Focus and Separation Network for Early Smoke Segmentation
ABSTRACT: Early smoke segmentation (ESS) enables the accurate identification of smoke
sources, facilitating the prompt extinguishing of fires and preventing
large-scale gas leaks. But ESS poses greater challenges than conventional
object and regular smoke segmentation due to its small scale and transparent
appearance, which can result in high miss detection rate and low precision. To
address these issues, a Focus and Separation Network (FoSp) is proposed. We
first introduce a Focus module employing bidirectional cascade which guides
low-resolution and high-resolution features towards mid-resolution to locate
and determine the scope of smoke, reducing the miss detection rate. Next, we
propose a Separation module that separates smoke images into a pure smoke
foreground and a smoke-free background, enhancing the contrast between smoke
and background fundamentally, improving segmentation precision. Finally, a
Domain Fusion module is developed to integrate the distinctive features of the
two modules which can balance recall and precision to achieve high F_beta.
Futhermore, to promote the development of ESS, we introduce a high-quality
real-world dataset called SmokeSeg, which contains more small and transparent
smoke than the existing datasets. Experimental results show that our model
achieves the best performance on three available datasets: SYN70K (mIoU:
83.00%), SMOKE5K (F_beta: 81.6%) and SmokeSeg (F_beta: 72.05%). Especially, our
FoSp outperforms SegFormer by 7.71% (F_beta) for early smoke segmentation on
SmokeSeg.
|
2306.04479 | Haiyang Liu | Haiyang Liu, Yuqi Fan, Lin Feng and Zhenchun Wei | Vulnerable Smart Contract Function Locating Based on Multi-Relational
Nested Graph Convolutional Network | Accepted By Jssoftware On JUNE 6, 2023 | null | null | null | cs.CR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The immutable and trustable characteristics of blockchain enable smart
contracts to be applied in various fields. Unfortunately, smart contracts are
subject to various vulnerabilities, which are frequently exploited by
attackers, causing financial damage to users.In this paper, we study the
problem of vulnerable smart contract function locating. We construct a novel
Multi-Relational Nested contract Graph (MRNG) to better characterize the rich
syntactic and semantic information in the smart contract code, including the
relationships between data and instructions. An MRNG represents a smart
contract, where each node represents a function in the smart contract and each
edge describes the calling relationship between the functions. In addition, we
create a Multi-Relational Function Graph (MRFG) for each function, which
characterizes the corresponding function code. That is, each function is
characterized as an MRFG, which corresponds to a node in the MRNG. Each MRFG
uses different types of edges to represent the different control and data
relationships between nodes within a function. We also propose a
Multi-Relational Nested Graph Convolutional Network (MRN-GCN) to process the
MRNG. MRN-GCN first extracts and aggregates features from each MRFG, using the
edge-enhanced graph convolution network and self-attention mechanism. The
extracted feature vector is then assigned to the corresponding node in the MRNG
to obtain a new Featured Contract Graph (FCG) for the smart contract. Graph
convolution is used to further extract features from the FCG. Finally, a feed
forward network with a Sigmoid function is used to locate the vulnerable
functions. Experimental results on the real-world smart contract datasets show
that model MRN-GCN can effectively improve the accuracy, precision, recall and
F1-score performance of vulnerable smart contract function locating.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 14:48:46 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Liu",
"Haiyang",
""
],
[
"Fan",
"Yuqi",
""
],
[
"Feng",
"Lin",
""
],
[
"Wei",
"Zhenchun",
""
]
] | TITLE: Vulnerable Smart Contract Function Locating Based on Multi-Relational
Nested Graph Convolutional Network
ABSTRACT: The immutable and trustable characteristics of blockchain enable smart
contracts to be applied in various fields. Unfortunately, smart contracts are
subject to various vulnerabilities, which are frequently exploited by
attackers, causing financial damage to users.In this paper, we study the
problem of vulnerable smart contract function locating. We construct a novel
Multi-Relational Nested contract Graph (MRNG) to better characterize the rich
syntactic and semantic information in the smart contract code, including the
relationships between data and instructions. An MRNG represents a smart
contract, where each node represents a function in the smart contract and each
edge describes the calling relationship between the functions. In addition, we
create a Multi-Relational Function Graph (MRFG) for each function, which
characterizes the corresponding function code. That is, each function is
characterized as an MRFG, which corresponds to a node in the MRNG. Each MRFG
uses different types of edges to represent the different control and data
relationships between nodes within a function. We also propose a
Multi-Relational Nested Graph Convolutional Network (MRN-GCN) to process the
MRNG. MRN-GCN first extracts and aggregates features from each MRFG, using the
edge-enhanced graph convolution network and self-attention mechanism. The
extracted feature vector is then assigned to the corresponding node in the MRNG
to obtain a new Featured Contract Graph (FCG) for the smart contract. Graph
convolution is used to further extract features from the FCG. Finally, a feed
forward network with a Sigmoid function is used to locate the vulnerable
functions. Experimental results on the real-world smart contract datasets show
that model MRN-GCN can effectively improve the accuracy, precision, recall and
F1-score performance of vulnerable smart contract function locating.
|
2306.04482 | Sruthi Sudhakar | Sruthi Sudhakar, Viraj Prabhu, Olga Russakovsky, Judy Hoffman | ICON$^2$: Reliably Benchmarking Predictive Inequity in Object Detection | Accepted to CVPR 2023 SSAD Workshop | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As computer vision systems are being increasingly deployed at scale in
high-stakes applications like autonomous driving, concerns about social bias in
these systems are rising. Analysis of fairness in real-world vision systems,
such as object detection in driving scenes, has been limited to observing
predictive inequity across attributes such as pedestrian skin tone, and lacks a
consistent methodology to disentangle the role of confounding variables e.g.
does my model perform worse for a certain skin tone, or are such scenes in my
dataset more challenging due to occlusion and crowds? In this work, we
introduce ICON$^2$, a framework for robustly answering this question. ICON$^2$
leverages prior knowledge on the deficiencies of object detection systems to
identify performance discrepancies across sub-populations, compute correlations
between these potential confounders and a given sensitive attribute, and
control for the most likely confounders to obtain a more reliable estimate of
model bias. Using our approach, we conduct an in-depth study on the performance
of object detection with respect to income from the BDD100K driving dataset,
revealing useful insights.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 17:42:42 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Sudhakar",
"Sruthi",
""
],
[
"Prabhu",
"Viraj",
""
],
[
"Russakovsky",
"Olga",
""
],
[
"Hoffman",
"Judy",
""
]
] | TITLE: ICON$^2$: Reliably Benchmarking Predictive Inequity in Object Detection
ABSTRACT: As computer vision systems are being increasingly deployed at scale in
high-stakes applications like autonomous driving, concerns about social bias in
these systems are rising. Analysis of fairness in real-world vision systems,
such as object detection in driving scenes, has been limited to observing
predictive inequity across attributes such as pedestrian skin tone, and lacks a
consistent methodology to disentangle the role of confounding variables e.g.
does my model perform worse for a certain skin tone, or are such scenes in my
dataset more challenging due to occlusion and crowds? In this work, we
introduce ICON$^2$, a framework for robustly answering this question. ICON$^2$
leverages prior knowledge on the deficiencies of object detection systems to
identify performance discrepancies across sub-populations, compute correlations
between these potential confounders and a given sensitive attribute, and
control for the most likely confounders to obtain a more reliable estimate of
model bias. Using our approach, we conduct an in-depth study on the performance
of object detection with respect to income from the BDD100K driving dataset,
revealing useful insights.
|
2306.04485 | Spencer Folk | Spencer Folk, James Paulos, Vijay Kumar | RotorPy: A Python-based Multirotor Simulator with Aerodynamics for
Education and Research | Appearing as a contributed paper in "The Role of Robotics Simulators
for Unmanned Aerial Vehicles" workshop at the 2023 International Conference
on Robotics and Automation (ICRA). See more at
https://imrclab.github.io/workshop-uav-sims-icra2023/ | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Simulators play a critical role in aerial robotics both in and out of the
classroom. We present RotorPy, a simulation environment written entirely in
Python intentionally designed to be a lightweight and accessible tool for
robotics students and researchers alike to probe concepts in estimation,
planning, and control for aerial robots. RotorPy simulates the 6-DoF dynamics
of a multirotor robot including aerodynamic wrenches, obstacles, actuator
dynamics and saturation, realistic sensors, and wind models. This work
describes the modeling choices for RotorPy, benchmark testing against real
data, and a case study using the simulator to design and evaluate a model-based
wind estimator.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 14:55:00 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Folk",
"Spencer",
""
],
[
"Paulos",
"James",
""
],
[
"Kumar",
"Vijay",
""
]
] | TITLE: RotorPy: A Python-based Multirotor Simulator with Aerodynamics for
Education and Research
ABSTRACT: Simulators play a critical role in aerial robotics both in and out of the
classroom. We present RotorPy, a simulation environment written entirely in
Python intentionally designed to be a lightweight and accessible tool for
robotics students and researchers alike to probe concepts in estimation,
planning, and control for aerial robots. RotorPy simulates the 6-DoF dynamics
of a multirotor robot including aerodynamic wrenches, obstacles, actuator
dynamics and saturation, realistic sensors, and wind models. This work
describes the modeling choices for RotorPy, benchmark testing against real
data, and a case study using the simulator to design and evaluate a model-based
wind estimator.
|
2306.04505 | Stephan W\"aldchen | Stephan W\"aldchen | Hardness of Deceptive Certificate Selection | 15 pages, 3 figures | null | null | null | cs.LG cs.AI cs.CC cs.CR | http://creativecommons.org/licenses/by-sa/4.0/ | Recent progress towards theoretical interpretability guarantees for AI has
been made with classifiers that are based on interactive proof systems. A
prover selects a certificate from the datapoint and sends it to a verifier who
decides the class. In the context of machine learning, such a certificate can
be a feature that is informative of the class. For a setup with high soundness
and completeness, the exchanged certificates must have a high mutual
information with the true class of the datapoint. However, this guarantee
relies on a bound on the Asymmetric Feature Correlation of the dataset, a
property that so far is difficult to estimate for high-dimensional data. It was
conjectured in W\"aldchen et al. that it is computationally hard to exploit the
AFC, which is what we prove here.
We consider a malicious prover-verifier duo that aims to exploit the AFC to
achieve high completeness and soundness while using uninformative certificates.
We show that this task is $\mathsf{NP}$-hard and cannot be approximated better
than $\mathcal{O}(m^{1/8 - \epsilon})$, where $m$ is the number of possible
certificates, for $\epsilon>0$ under the Dense-vs-Random conjecture. This is
some evidence that AFC should not prevent the use of interactive classification
for real-world tasks, as it is computationally hard to be exploited.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 15:12:16 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Wäldchen",
"Stephan",
""
]
] | TITLE: Hardness of Deceptive Certificate Selection
ABSTRACT: Recent progress towards theoretical interpretability guarantees for AI has
been made with classifiers that are based on interactive proof systems. A
prover selects a certificate from the datapoint and sends it to a verifier who
decides the class. In the context of machine learning, such a certificate can
be a feature that is informative of the class. For a setup with high soundness
and completeness, the exchanged certificates must have a high mutual
information with the true class of the datapoint. However, this guarantee
relies on a bound on the Asymmetric Feature Correlation of the dataset, a
property that so far is difficult to estimate for high-dimensional data. It was
conjectured in W\"aldchen et al. that it is computationally hard to exploit the
AFC, which is what we prove here.
We consider a malicious prover-verifier duo that aims to exploit the AFC to
achieve high completeness and soundness while using uninformative certificates.
We show that this task is $\mathsf{NP}$-hard and cannot be approximated better
than $\mathcal{O}(m^{1/8 - \epsilon})$, where $m$ is the number of possible
certificates, for $\epsilon>0$ under the Dense-vs-Random conjecture. This is
some evidence that AFC should not prevent the use of interactive classification
for real-world tasks, as it is computationally hard to be exploited.
|
2306.04508 | Gong Cheng | Zixian Huang, Jiaying Zhou, Gengyang Xiao, Gong Cheng | Enhancing In-Context Learning with Answer Feedback for Multi-Span
Question Answering | 12 pages, submitted to NLPCC 2023 | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Whereas the recent emergence of large language models (LLMs) like ChatGPT has
exhibited impressive general performance, it still has a large gap with
fully-supervised models on specific tasks such as multi-span question
answering. Previous researches found that in-context learning is an effective
approach to exploiting LLM, by using a few task-related labeled data as
demonstration examples to construct a few-shot prompt for answering new
questions. A popular implementation is to concatenate a few questions and their
correct answers through simple templates, informing LLM of the desired output.
In this paper, we propose a novel way of employing labeled data such that it
also informs LLM of some undesired output, by extending demonstration examples
with feedback about answers predicted by an off-the-shelf model, e.g., correct,
incorrect, or incomplete. Experiments on three multi-span question answering
datasets as well as a keyphrase extraction dataset show that our new prompting
strategy consistently improves LLM's in-context learning performance.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 15:20:24 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Huang",
"Zixian",
""
],
[
"Zhou",
"Jiaying",
""
],
[
"Xiao",
"Gengyang",
""
],
[
"Cheng",
"Gong",
""
]
] | TITLE: Enhancing In-Context Learning with Answer Feedback for Multi-Span
Question Answering
ABSTRACT: Whereas the recent emergence of large language models (LLMs) like ChatGPT has
exhibited impressive general performance, it still has a large gap with
fully-supervised models on specific tasks such as multi-span question
answering. Previous researches found that in-context learning is an effective
approach to exploiting LLM, by using a few task-related labeled data as
demonstration examples to construct a few-shot prompt for answering new
questions. A popular implementation is to concatenate a few questions and their
correct answers through simple templates, informing LLM of the desired output.
In this paper, we propose a novel way of employing labeled data such that it
also informs LLM of some undesired output, by extending demonstration examples
with feedback about answers predicted by an off-the-shelf model, e.g., correct,
incorrect, or incomplete. Experiments on three multi-span question answering
datasets as well as a keyphrase extraction dataset show that our new prompting
strategy consistently improves LLM's in-context learning performance.
|
2306.04519 | Emilie Gr\'egoire | Emilie Gr\'egoire, Hafeez Chaudhary and Sam Verboven | Sample-Level Weighting for Multi-Task Learning with Auxiliary Tasks | 16 pages, 7 figures | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multi-task learning (MTL) can improve the generalization performance of
neural networks by sharing representations with related tasks. Nonetheless, MTL
can also degrade performance through harmful interference between tasks. Recent
work has pursued task-specific loss weighting as a solution for this
interference. However, existing algorithms treat tasks as atomic, lacking the
ability to explicitly separate harmful and helpful signals beyond the task
level. To this end, we propose SLGrad, a sample-level weighting algorithm for
multi-task learning with auxiliary tasks. Through sample-specific task weights,
SLGrad reshapes the task distributions during training to eliminate harmful
auxiliary signals and augment useful task signals. Substantial generalization
performance gains are observed on (semi-) synthetic datasets and common
supervised multi-task problems.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 15:29:46 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Grégoire",
"Emilie",
""
],
[
"Chaudhary",
"Hafeez",
""
],
[
"Verboven",
"Sam",
""
]
] | TITLE: Sample-Level Weighting for Multi-Task Learning with Auxiliary Tasks
ABSTRACT: Multi-task learning (MTL) can improve the generalization performance of
neural networks by sharing representations with related tasks. Nonetheless, MTL
can also degrade performance through harmful interference between tasks. Recent
work has pursued task-specific loss weighting as a solution for this
interference. However, existing algorithms treat tasks as atomic, lacking the
ability to explicitly separate harmful and helpful signals beyond the task
level. To this end, we propose SLGrad, a sample-level weighting algorithm for
multi-task learning with auxiliary tasks. Through sample-specific task weights,
SLGrad reshapes the task distributions during training to eliminate harmful
auxiliary signals and augment useful task signals. Substantial generalization
performance gains are observed on (semi-) synthetic datasets and common
supervised multi-task problems.
|
2306.04523 | Ines Reinig | Ines Reinig and Katja Markert | Can current NLI systems handle German word order? Investigating language
model performance on a new German challenge set of minimal pairs | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Compared to English, German word order is freer and therefore poses
additional challenges for natural language inference (NLI). We create WOGLI
(Word Order in German Language Inference), the first adversarial NLI dataset
for German word order that has the following properties: (i) each premise has
an entailed and a non-entailed hypothesis; (ii) premise and hypotheses differ
only in word order and necessary morphological changes to mark case and number.
In particular, each premise andits two hypotheses contain exactly the same
lemmata. Our adversarial examples require the model to use morphological
markers in order to recognise or reject entailment. We show that current German
autoencoding models fine-tuned on translated NLI data can struggle on this
challenge set, reflecting the fact that translated NLI datasets will not mirror
all necessary language phenomena in the target language. We also examine
performance after data augmentation as well as on related word order phenomena
derived from WOGLI. Our datasets are publically available at
https://github.com/ireinig/wogli.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 15:33:07 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Reinig",
"Ines",
""
],
[
"Markert",
"Katja",
""
]
] | TITLE: Can current NLI systems handle German word order? Investigating language
model performance on a new German challenge set of minimal pairs
ABSTRACT: Compared to English, German word order is freer and therefore poses
additional challenges for natural language inference (NLI). We create WOGLI
(Word Order in German Language Inference), the first adversarial NLI dataset
for German word order that has the following properties: (i) each premise has
an entailed and a non-entailed hypothesis; (ii) premise and hypotheses differ
only in word order and necessary morphological changes to mark case and number.
In particular, each premise andits two hypotheses contain exactly the same
lemmata. Our adversarial examples require the model to use morphological
markers in order to recognise or reject entailment. We show that current German
autoencoding models fine-tuned on translated NLI data can struggle on this
challenge set, reflecting the fact that translated NLI datasets will not mirror
all necessary language phenomena in the target language. We also examine
performance after data augmentation as well as on related word order phenomena
derived from WOGLI. Our datasets are publically available at
https://github.com/ireinig/wogli.
|
2306.04540 | Xi Zhu | Xi Zhu, Xiya Cao, Zhiwei Dong, Caifa Zhou, Qiangbo Liu, Wei Li,
Yongliang Wang | NeMO: Neural Map Growing System for Spatiotemporal Fusion in
Bird's-Eye-View and BDD-Map Benchmark | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Vision-centric Bird's-Eye View (BEV) representation is essential for
autonomous driving systems (ADS). Multi-frame temporal fusion which leverages
historical information has been demonstrated to provide more comprehensive
perception results. While most research focuses on ego-centric maps of fixed
settings, long-range local map generation remains less explored. This work
outlines a new paradigm, named NeMO, for generating local maps through the
utilization of a readable and writable big map, a learning-based fusion module,
and an interaction mechanism between the two. With an assumption that the
feature distribution of all BEV grids follows an identical pattern, we adopt a
shared-weight neural network for all grids to update the big map. This paradigm
supports the fusion of longer time series and the generation of long-range BEV
local maps. Furthermore, we release BDD-Map, a BDD100K-based dataset
incorporating map element annotations, including lane lines, boundaries, and
pedestrian crossing. Experiments on the NuScenes and BDD-Map datasets
demonstrate that NeMO outperforms state-of-the-art map segmentation methods. We
also provide a new scene-level BEV map evaluation setting along with the
corresponding baseline for a more comprehensive comparison.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 15:46:15 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Zhu",
"Xi",
""
],
[
"Cao",
"Xiya",
""
],
[
"Dong",
"Zhiwei",
""
],
[
"Zhou",
"Caifa",
""
],
[
"Liu",
"Qiangbo",
""
],
[
"Li",
"Wei",
""
],
[
"Wang",
"Yongliang",
""
]
] | TITLE: NeMO: Neural Map Growing System for Spatiotemporal Fusion in
Bird's-Eye-View and BDD-Map Benchmark
ABSTRACT: Vision-centric Bird's-Eye View (BEV) representation is essential for
autonomous driving systems (ADS). Multi-frame temporal fusion which leverages
historical information has been demonstrated to provide more comprehensive
perception results. While most research focuses on ego-centric maps of fixed
settings, long-range local map generation remains less explored. This work
outlines a new paradigm, named NeMO, for generating local maps through the
utilization of a readable and writable big map, a learning-based fusion module,
and an interaction mechanism between the two. With an assumption that the
feature distribution of all BEV grids follows an identical pattern, we adopt a
shared-weight neural network for all grids to update the big map. This paradigm
supports the fusion of longer time series and the generation of long-range BEV
local maps. Furthermore, we release BDD-Map, a BDD100K-based dataset
incorporating map element annotations, including lane lines, boundaries, and
pedestrian crossing. Experiments on the NuScenes and BDD-Map datasets
demonstrate that NeMO outperforms state-of-the-art map segmentation methods. We
also provide a new scene-level BEV map evaluation setting along with the
corresponding baseline for a more comprehensive comparison.
|
2306.04556 | Arjun Guha | Hannah McLean Babe, Sydney Nguyen, Yangtian Zi, Arjun Guha, Molly Q
Feldman, Carolyn Jane Anderson | StudentEval: A Benchmark of Student-Written Prompts for Large Language
Models of Code | null | null | null | null | cs.LG cs.HC cs.SE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Code LLMs are being rapidly deployed and there is evidence that they can make
professional programmers more productive. Current benchmarks for code
generation measure whether models generate correct programs given an expert
prompt. In this paper, we present a new benchmark containing multiple prompts
per problem, written by a specific population of non-expert prompters:
beginning programmers. StudentEval contains 1,749 prompts for 48 problems,
written by 80 students who have only completed one semester of Python
programming. Our students wrote these prompts while working interactively with
a Code LLM, and we observed very mixed success rates. We use StudentEval to
evaluate 5 Code LLMs and find that StudentEval is a better discriminator of
model performance than existing benchmarks. We analyze the prompts and find
significant variation in students' prompting techniques. We also find that
nondeterministic LLM sampling could mislead students into thinking that their
prompts are more (or less) effective than they actually are, which has
implications for how to teach with Code LLMs.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 16:03:55 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Babe",
"Hannah McLean",
""
],
[
"Nguyen",
"Sydney",
""
],
[
"Zi",
"Yangtian",
""
],
[
"Guha",
"Arjun",
""
],
[
"Feldman",
"Molly Q",
""
],
[
"Anderson",
"Carolyn Jane",
""
]
] | TITLE: StudentEval: A Benchmark of Student-Written Prompts for Large Language
Models of Code
ABSTRACT: Code LLMs are being rapidly deployed and there is evidence that they can make
professional programmers more productive. Current benchmarks for code
generation measure whether models generate correct programs given an expert
prompt. In this paper, we present a new benchmark containing multiple prompts
per problem, written by a specific population of non-expert prompters:
beginning programmers. StudentEval contains 1,749 prompts for 48 problems,
written by 80 students who have only completed one semester of Python
programming. Our students wrote these prompts while working interactively with
a Code LLM, and we observed very mixed success rates. We use StudentEval to
evaluate 5 Code LLMs and find that StudentEval is a better discriminator of
model performance than existing benchmarks. We analyze the prompts and find
significant variation in students' prompting techniques. We also find that
nondeterministic LLM sampling could mislead students into thinking that their
prompts are more (or less) effective than they actually are, which has
implications for how to teach with Code LLMs.
|
2306.04579 | Kaidong Zhang | Kaidong Zhang, Ziyang Gan, Dong Liu, Xifu Shang | A Dataset for Deep Learning-based Bone Structure Analyses in Total Hip
Arthroplasty | 16 pages, 17 figures | null | null | null | eess.IV cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Total hip arthroplasty (THA) is a widely used surgical procedure in
orthopedics. For THA, it is of clinical significance to analyze the bone
structure from the CT images, especially to observe the structure of the
acetabulum and femoral head, before the surgical procedure. For such bone
structure analyses, deep learning technologies are promising but require
high-quality labeled data for the learning, while the data labeling is costly.
We address this issue and propose an efficient data annotation pipeline for
producing a deep learning-oriented dataset. Our pipeline consists of
non-learning-based bone extraction (BE) and acetabulum and femoral head
segmentation (AFS) and active-learning-based annotation refinement (AAR). For
BE we use the classic graph-cut algorithm. For AFS we propose an improved
algorithm, including femoral head boundary localization using first-order and
second-order gradient regularization, line-based non-maximum suppression, and
anatomy prior-based femoral head extraction. For AAR, we refine the
algorithm-produced pseudo labels with the help of trained deep models: we
measure the uncertainty based on the disagreement between the original pseudo
labels and the deep model predictions, and then find out the samples with the
largest uncertainty to ask for manual labeling. Using the proposed pipeline, we
construct a large-scale bone structure analyses dataset from more than 300
clinical and diverse CT scans. We perform careful manual labeling for the test
set of our data. We then benchmark multiple state-of-the art deep
learning-based methods of medical image segmentation using the training and
test sets of our data. The extensive experimental results validate the efficacy
of the proposed data annotation pipeline. The dataset, related codes and models
will be publicly available at https://github.com/hitachinsk/THA.
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 16:28:53 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Zhang",
"Kaidong",
""
],
[
"Gan",
"Ziyang",
""
],
[
"Liu",
"Dong",
""
],
[
"Shang",
"Xifu",
""
]
] | TITLE: A Dataset for Deep Learning-based Bone Structure Analyses in Total Hip
Arthroplasty
ABSTRACT: Total hip arthroplasty (THA) is a widely used surgical procedure in
orthopedics. For THA, it is of clinical significance to analyze the bone
structure from the CT images, especially to observe the structure of the
acetabulum and femoral head, before the surgical procedure. For such bone
structure analyses, deep learning technologies are promising but require
high-quality labeled data for the learning, while the data labeling is costly.
We address this issue and propose an efficient data annotation pipeline for
producing a deep learning-oriented dataset. Our pipeline consists of
non-learning-based bone extraction (BE) and acetabulum and femoral head
segmentation (AFS) and active-learning-based annotation refinement (AAR). For
BE we use the classic graph-cut algorithm. For AFS we propose an improved
algorithm, including femoral head boundary localization using first-order and
second-order gradient regularization, line-based non-maximum suppression, and
anatomy prior-based femoral head extraction. For AAR, we refine the
algorithm-produced pseudo labels with the help of trained deep models: we
measure the uncertainty based on the disagreement between the original pseudo
labels and the deep model predictions, and then find out the samples with the
largest uncertainty to ask for manual labeling. Using the proposed pipeline, we
construct a large-scale bone structure analyses dataset from more than 300
clinical and diverse CT scans. We perform careful manual labeling for the test
set of our data. We then benchmark multiple state-of-the art deep
learning-based methods of medical image segmentation using the training and
test sets of our data. The extensive experimental results validate the efficacy
of the proposed data annotation pipeline. The dataset, related codes and models
will be publicly available at https://github.com/hitachinsk/THA.
|
2306.04610 | Nicholas Riccardi | Nicholas Riccardi and Rutvik H. Desai | The Two Word Test: A Semantic Benchmark for Large Language Models | 12 pages, 5 figures, 3 tables, submitted to NeurIPS 2023 Datasets and
Benchmarks Track | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Large Language Models (LLMs) have shown remarkable abilities recently,
including passing advanced professional exams and demanding benchmark tests.
This performance has led many to suggest that they are close to achieving
humanlike or 'true' understanding of language, and even Artificial General
Intelligence (AGI). Here, we provide a new open-source benchmark that can
assess semantic abilities of LLMs using two-word phrases using a task that can
be performed relatively easily by humans without advanced training. Combining
multiple words into a single concept is a fundamental aspect of human language
and intelligence. The test requires meaningfulness judgments of 1768 noun-noun
combinations that have been rated as meaningful (e.g., baby boy) or not
meaningful (e.g., goat sky). by 150 human raters. We provide versions of the
task that probe meaningfulness ratings on a 0-4 scale as well as binary
judgments. We conducted a series of experiments using the TWT on GPT-4,
GPT-3.5, and Bard, with both versions. Results demonstrated that, compared to
humans, all models perform poorly at rating meaningfulness of these phrases.
GPT-3.5 and Bard are also unable to make binary discriminations between
sensible and nonsense phrases as making sense. GPT-4 makes a substantial
improvement in binary discrimination of combinatorial phrases but is still
significantly worse than human performance. The TWT can be used to understand
the limitations and weaknesses of current LLMs, and potentially improve them.
The test also reminds us that caution is warranted in attributing 'true
understanding' or AGI to LLMs. TWT is available at:
https://github.com/NickRiccardi/two-word-test
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 17:22:03 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Riccardi",
"Nicholas",
""
],
[
"Desai",
"Rutvik H.",
""
]
] | TITLE: The Two Word Test: A Semantic Benchmark for Large Language Models
ABSTRACT: Large Language Models (LLMs) have shown remarkable abilities recently,
including passing advanced professional exams and demanding benchmark tests.
This performance has led many to suggest that they are close to achieving
humanlike or 'true' understanding of language, and even Artificial General
Intelligence (AGI). Here, we provide a new open-source benchmark that can
assess semantic abilities of LLMs using two-word phrases using a task that can
be performed relatively easily by humans without advanced training. Combining
multiple words into a single concept is a fundamental aspect of human language
and intelligence. The test requires meaningfulness judgments of 1768 noun-noun
combinations that have been rated as meaningful (e.g., baby boy) or not
meaningful (e.g., goat sky). by 150 human raters. We provide versions of the
task that probe meaningfulness ratings on a 0-4 scale as well as binary
judgments. We conducted a series of experiments using the TWT on GPT-4,
GPT-3.5, and Bard, with both versions. Results demonstrated that, compared to
humans, all models perform poorly at rating meaningfulness of these phrases.
GPT-3.5 and Bard are also unable to make binary discriminations between
sensible and nonsense phrases as making sense. GPT-4 makes a substantial
improvement in binary discrimination of combinatorial phrases but is still
significantly worse than human performance. The TWT can be used to understand
the limitations and weaknesses of current LLMs, and potentially improve them.
The test also reminds us that caution is warranted in attributing 'true
understanding' or AGI to LLMs. TWT is available at:
https://github.com/NickRiccardi/two-word-test
|
2306.04619 | Chun-Han Yao | Chun-Han Yao, Amit Raj, Wei-Chih Hung, Yuanzhen Li, Michael
Rubinstein, Ming-Hsuan Yang, Varun Jampani | ARTIC3D: Learning Robust Articulated 3D Shapes from Noisy Web Image
Collections | Project page: https://chhankyao.github.io/artic3d/ | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Estimating 3D articulated shapes like animal bodies from monocular images is
inherently challenging due to the ambiguities of camera viewpoint, pose,
texture, lighting, etc. We propose ARTIC3D, a self-supervised framework to
reconstruct per-instance 3D shapes from a sparse image collection in-the-wild.
Specifically, ARTIC3D is built upon a skeleton-based surface representation and
is further guided by 2D diffusion priors from Stable Diffusion. First, we
enhance the input images with occlusions/truncation via 2D diffusion to obtain
cleaner mask estimates and semantic features. Second, we perform
diffusion-guided 3D optimization to estimate shape and texture that are of
high-fidelity and faithful to input images. We also propose a novel technique
to calculate more stable image-level gradients via diffusion models compared to
existing alternatives. Finally, we produce realistic animations by fine-tuning
the rendered shape and texture under rigid part transformations. Extensive
evaluations on multiple existing datasets as well as newly introduced noisy web
image collections with occlusions and truncation demonstrate that ARTIC3D
outputs are more robust to noisy images, higher quality in terms of shape and
texture details, and more realistic when animated. Project page:
https://chhankyao.github.io/artic3d/
| [
{
"version": "v1",
"created": "Wed, 7 Jun 2023 17:47:50 GMT"
}
] | 2023-06-08T00:00:00 | [
[
"Yao",
"Chun-Han",
""
],
[
"Raj",
"Amit",
""
],
[
"Hung",
"Wei-Chih",
""
],
[
"Li",
"Yuanzhen",
""
],
[
"Rubinstein",
"Michael",
""
],
[
"Yang",
"Ming-Hsuan",
""
],
[
"Jampani",
"Varun",
""
]
] | TITLE: ARTIC3D: Learning Robust Articulated 3D Shapes from Noisy Web Image
Collections
ABSTRACT: Estimating 3D articulated shapes like animal bodies from monocular images is
inherently challenging due to the ambiguities of camera viewpoint, pose,
texture, lighting, etc. We propose ARTIC3D, a self-supervised framework to
reconstruct per-instance 3D shapes from a sparse image collection in-the-wild.
Specifically, ARTIC3D is built upon a skeleton-based surface representation and
is further guided by 2D diffusion priors from Stable Diffusion. First, we
enhance the input images with occlusions/truncation via 2D diffusion to obtain
cleaner mask estimates and semantic features. Second, we perform
diffusion-guided 3D optimization to estimate shape and texture that are of
high-fidelity and faithful to input images. We also propose a novel technique
to calculate more stable image-level gradients via diffusion models compared to
existing alternatives. Finally, we produce realistic animations by fine-tuning
the rendered shape and texture under rigid part transformations. Extensive
evaluations on multiple existing datasets as well as newly introduced noisy web
image collections with occlusions and truncation demonstrate that ARTIC3D
outputs are more robust to noisy images, higher quality in terms of shape and
texture details, and more realistic when animated. Project page:
https://chhankyao.github.io/artic3d/
|
2007.06676 | Senthil Yogamani | Varun Ravi Kumar, Senthil Yogamani, Markus Bach, Christian Witt,
Stefan Milz and Patrick Mader | UnRectDepthNet: Self-Supervised Monocular Depth Estimation using a
Generic Framework for Handling Common Camera Distortion Models | Minor fixes added after IROS 2020 Camera ready submission. IROS 2020
presentation video - https://www.youtube.com/watch?v=3Br2KSWZRrY | null | null | null | cs.CV cs.LG cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In classical computer vision, rectification is an integral part of multi-view
depth estimation. It typically includes epipolar rectification and lens
distortion correction. This process simplifies the depth estimation
significantly, and thus it has been adopted in CNN approaches. However,
rectification has several side effects, including a reduced field of view
(FOV), resampling distortion, and sensitivity to calibration errors. The
effects are particularly pronounced in case of significant distortion (e.g.,
wide-angle fisheye cameras). In this paper, we propose a generic scale-aware
self-supervised pipeline for estimating depth, euclidean distance, and visual
odometry from unrectified monocular videos. We demonstrate a similar level of
precision on the unrectified KITTI dataset with barrel distortion comparable to
the rectified KITTI dataset. The intuition being that the rectification step
can be implicitly absorbed within the CNN model, which learns the distortion
model without increasing complexity. Our approach does not suffer from a
reduced field of view and avoids computational costs for rectification at
inference time. To further illustrate the general applicability of the proposed
framework, we apply it to wide-angle fisheye cameras with 190$^\circ$
horizontal field of view. The training framework UnRectDepthNet takes in the
camera distortion model as an argument and adapts projection and unprojection
functions accordingly. The proposed algorithm is evaluated further on the KITTI
rectified dataset, and we achieve state-of-the-art results that improve upon
our previous work FisheyeDistanceNet. Qualitative results on a distorted test
scene video sequence indicate excellent performance
https://youtu.be/K6pbx3bU4Ss.
| [
{
"version": "v1",
"created": "Mon, 13 Jul 2020 20:35:05 GMT"
},
{
"version": "v2",
"created": "Sun, 26 Jul 2020 14:04:24 GMT"
},
{
"version": "v3",
"created": "Tue, 6 Oct 2020 19:41:35 GMT"
},
{
"version": "v4",
"created": "Tue, 6 Jun 2023 14:26:28 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Kumar",
"Varun Ravi",
""
],
[
"Yogamani",
"Senthil",
""
],
[
"Bach",
"Markus",
""
],
[
"Witt",
"Christian",
""
],
[
"Milz",
"Stefan",
""
],
[
"Mader",
"Patrick",
""
]
] | TITLE: UnRectDepthNet: Self-Supervised Monocular Depth Estimation using a
Generic Framework for Handling Common Camera Distortion Models
ABSTRACT: In classical computer vision, rectification is an integral part of multi-view
depth estimation. It typically includes epipolar rectification and lens
distortion correction. This process simplifies the depth estimation
significantly, and thus it has been adopted in CNN approaches. However,
rectification has several side effects, including a reduced field of view
(FOV), resampling distortion, and sensitivity to calibration errors. The
effects are particularly pronounced in case of significant distortion (e.g.,
wide-angle fisheye cameras). In this paper, we propose a generic scale-aware
self-supervised pipeline for estimating depth, euclidean distance, and visual
odometry from unrectified monocular videos. We demonstrate a similar level of
precision on the unrectified KITTI dataset with barrel distortion comparable to
the rectified KITTI dataset. The intuition being that the rectification step
can be implicitly absorbed within the CNN model, which learns the distortion
model without increasing complexity. Our approach does not suffer from a
reduced field of view and avoids computational costs for rectification at
inference time. To further illustrate the general applicability of the proposed
framework, we apply it to wide-angle fisheye cameras with 190$^\circ$
horizontal field of view. The training framework UnRectDepthNet takes in the
camera distortion model as an argument and adapts projection and unprojection
functions accordingly. The proposed algorithm is evaluated further on the KITTI
rectified dataset, and we achieve state-of-the-art results that improve upon
our previous work FisheyeDistanceNet. Qualitative results on a distorted test
scene video sequence indicate excellent performance
https://youtu.be/K6pbx3bU4Ss.
|
2011.11062 | Eduardo Bezerra | Marcello Serqueira, Pedro Gonz\'alez, Eduardo Bezerra | A Population-based Hybrid Approach to Hyperparameter Optimization for
Neural Networks | 28 pages, 7 figures | null | 10.1109/ACCESS.2023.3277310 | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | In recent years, large amounts of data have been generated, and computer
power has kept growing. This scenario has led to a resurgence in the interest
in artificial neural networks. One of the main challenges in training effective
neural network models is finding the right combination of hyperparameters to be
used. Indeed, the choice of an adequate approach to search the hyperparameter
space directly influences the accuracy of the resulting neural network model.
Common approaches for hyperparameter optimization are Grid Search, Random
Search, and Bayesian Optimization. There are also population-based methods such
as CMA-ES. In this paper, we present HBRKGA, a new population-based approach
for hyperparameter optimization. HBRKGA is a hybrid approach that combines the
Biased Random Key Genetic Algorithm with a Random Walk technique to search the
hyperparameter space efficiently. Several computational experiments on eight
different datasets were performed to assess the effectiveness of the proposed
approach. Results showed that HBRKGA could find hyperparameter configurations
that outperformed (in terms of predictive quality) the baseline methods in six
out of eight datasets while showing a reasonable execution time.
| [
{
"version": "v1",
"created": "Sun, 22 Nov 2020 17:12:31 GMT"
},
{
"version": "v2",
"created": "Fri, 27 Nov 2020 12:58:40 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Serqueira",
"Marcello",
""
],
[
"González",
"Pedro",
""
],
[
"Bezerra",
"Eduardo",
""
]
] | TITLE: A Population-based Hybrid Approach to Hyperparameter Optimization for
Neural Networks
ABSTRACT: In recent years, large amounts of data have been generated, and computer
power has kept growing. This scenario has led to a resurgence in the interest
in artificial neural networks. One of the main challenges in training effective
neural network models is finding the right combination of hyperparameters to be
used. Indeed, the choice of an adequate approach to search the hyperparameter
space directly influences the accuracy of the resulting neural network model.
Common approaches for hyperparameter optimization are Grid Search, Random
Search, and Bayesian Optimization. There are also population-based methods such
as CMA-ES. In this paper, we present HBRKGA, a new population-based approach
for hyperparameter optimization. HBRKGA is a hybrid approach that combines the
Biased Random Key Genetic Algorithm with a Random Walk technique to search the
hyperparameter space efficiently. Several computational experiments on eight
different datasets were performed to assess the effectiveness of the proposed
approach. Results showed that HBRKGA could find hyperparameter configurations
that outperformed (in terms of predictive quality) the baseline methods in six
out of eight datasets while showing a reasonable execution time.
|
2102.07448 | Senthil Yogamani | Varun Ravi Kumar, Senthil Yogamani, Hazem Rashed, Ganesh Sistu,
Christian Witt, Isabelle Leang, Stefan Milz and Patrick M\"ader | OmniDet: Surround View Cameras based Multi-task Visual Perception
Network for Autonomous Driving | Best Robot Vision paper award finalist (top 4). Camera ready version
accepted for RA-L and ICRA 2021 publication | null | null | null | cs.CV cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Surround View fisheye cameras are commonly deployed in automated driving for
360\deg{} near-field sensing around the vehicle. This work presents a
multi-task visual perception network on unrectified fisheye images to enable
the vehicle to sense its surrounding environment. It consists of six primary
tasks necessary for an autonomous driving system: depth estimation, visual
odometry, semantic segmentation, motion segmentation, object detection, and
lens soiling detection. We demonstrate that the jointly trained model performs
better than the respective single task versions. Our multi-task model has a
shared encoder providing a significant computational advantage and has
synergized decoders where tasks support each other. We propose a novel camera
geometry based adaptation mechanism to encode the fisheye distortion model both
at training and inference. This was crucial to enable training on the WoodScape
dataset, comprised of data from different parts of the world collected by 12
different cameras mounted on three different cars with different intrinsics and
viewpoints. Given that bounding boxes is not a good representation for
distorted fisheye images, we also extend object detection to use a polygon with
non-uniformly sampled vertices. We additionally evaluate our model on standard
automotive datasets, namely KITTI and Cityscapes. We obtain the
state-of-the-art results on KITTI for depth estimation and pose estimation
tasks and competitive performance on the other tasks. We perform extensive
ablation studies on various architecture choices and task weighting
methodologies. A short video at https://youtu.be/xbSjZ5OfPes provides
qualitative results.
| [
{
"version": "v1",
"created": "Mon, 15 Feb 2021 10:46:24 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Aug 2021 14:45:16 GMT"
},
{
"version": "v3",
"created": "Tue, 6 Jun 2023 14:31:21 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Kumar",
"Varun Ravi",
""
],
[
"Yogamani",
"Senthil",
""
],
[
"Rashed",
"Hazem",
""
],
[
"Sistu",
"Ganesh",
""
],
[
"Witt",
"Christian",
""
],
[
"Leang",
"Isabelle",
""
],
[
"Milz",
"Stefan",
""
],
[
"Mäder",
"Patrick",
""
]
] | TITLE: OmniDet: Surround View Cameras based Multi-task Visual Perception
Network for Autonomous Driving
ABSTRACT: Surround View fisheye cameras are commonly deployed in automated driving for
360\deg{} near-field sensing around the vehicle. This work presents a
multi-task visual perception network on unrectified fisheye images to enable
the vehicle to sense its surrounding environment. It consists of six primary
tasks necessary for an autonomous driving system: depth estimation, visual
odometry, semantic segmentation, motion segmentation, object detection, and
lens soiling detection. We demonstrate that the jointly trained model performs
better than the respective single task versions. Our multi-task model has a
shared encoder providing a significant computational advantage and has
synergized decoders where tasks support each other. We propose a novel camera
geometry based adaptation mechanism to encode the fisheye distortion model both
at training and inference. This was crucial to enable training on the WoodScape
dataset, comprised of data from different parts of the world collected by 12
different cameras mounted on three different cars with different intrinsics and
viewpoints. Given that bounding boxes is not a good representation for
distorted fisheye images, we also extend object detection to use a polygon with
non-uniformly sampled vertices. We additionally evaluate our model on standard
automotive datasets, namely KITTI and Cityscapes. We obtain the
state-of-the-art results on KITTI for depth estimation and pose estimation
tasks and competitive performance on the other tasks. We perform extensive
ablation studies on various architecture choices and task weighting
methodologies. A short video at https://youtu.be/xbSjZ5OfPes provides
qualitative results.
|
2103.00268 | Naoki Wake | Naoki Wake, Daichi Saito, Kazuhiro Sasabuchi, Hideki Koike, Katsushi
Ikeuchi | Text-driven object affordance for guiding grasp-type recognition in
multimodal robot teaching | 8 pages, 11 figures. Last updated March 12, 2023 Accepted for
publication in Machine Vision and Applications | null | 10.1007/s00138-023-01408-z | null | cs.RO cs.CV cs.HC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This study investigates how text-driven object affordance, which provides
prior knowledge about grasp types for each object, affects image-based
grasp-type recognition in robot teaching. The researchers created labeled
datasets of first-person hand images to examine the impact of object affordance
on recognition performance. They evaluated scenarios with real and illusory
objects, considering mixed reality teaching conditions where visual object
information may be limited. The results demonstrate that object affordance
improves image-based recognition by filtering out unlikely grasp types and
emphasizing likely ones. The effectiveness of object affordance was more
pronounced when there was a stronger bias towards specific grasp types for each
object. These findings highlight the significance of object affordance in
multimodal robot teaching, regardless of whether real objects are present in
the images. Sample code is available on
https://github.com/microsoft/arr-grasp-type-recognition.
| [
{
"version": "v1",
"created": "Sat, 27 Feb 2021 17:03:32 GMT"
},
{
"version": "v2",
"created": "Fri, 12 May 2023 12:35:49 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Wake",
"Naoki",
""
],
[
"Saito",
"Daichi",
""
],
[
"Sasabuchi",
"Kazuhiro",
""
],
[
"Koike",
"Hideki",
""
],
[
"Ikeuchi",
"Katsushi",
""
]
] | TITLE: Text-driven object affordance for guiding grasp-type recognition in
multimodal robot teaching
ABSTRACT: This study investigates how text-driven object affordance, which provides
prior knowledge about grasp types for each object, affects image-based
grasp-type recognition in robot teaching. The researchers created labeled
datasets of first-person hand images to examine the impact of object affordance
on recognition performance. They evaluated scenarios with real and illusory
objects, considering mixed reality teaching conditions where visual object
information may be limited. The results demonstrate that object affordance
improves image-based recognition by filtering out unlikely grasp types and
emphasizing likely ones. The effectiveness of object affordance was more
pronounced when there was a stronger bias towards specific grasp types for each
object. These findings highlight the significance of object affordance in
multimodal robot teaching, regardless of whether real objects are present in
the images. Sample code is available on
https://github.com/microsoft/arr-grasp-type-recognition.
|
2202.00179 | Dong Huo | Dong Huo, Abbas Masoumzadeh, Rafsanjany Kushol, Yee-Hong Yang | Blind Image Deconvolution Using Variational Deep Image Prior | IEEE Transactions on Pattern Analysis and Machine Intelligence
(TPAMI), 2023 | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Conventional deconvolution methods utilize hand-crafted image priors to
constrain the optimization. While deep-learning-based methods have simplified
the optimization by end-to-end training, they fail to generalize well to blurs
unseen in the training dataset. Thus, training image-specific models is
important for higher generalization. Deep image prior (DIP) provides an
approach to optimize the weights of a randomly initialized network with a
single degraded image by maximum a posteriori (MAP), which shows that the
architecture of a network can serve as the hand-crafted image prior. Different
from the conventional hand-crafted image priors that are statistically
obtained, it is hard to find a proper network architecture because the
relationship between images and their corresponding network architectures is
unclear. As a result, the network architecture cannot provide enough constraint
for the latent sharp image. This paper proposes a new variational deep image
prior (VDIP) for blind image deconvolution, which exploits additive
hand-crafted image priors on latent sharp images and approximates a
distribution for each pixel to avoid suboptimal solutions. Our mathematical
analysis shows that the proposed method can better constrain the optimization.
The experimental results further demonstrate that the generated images have
better quality than that of the original DIP on benchmark datasets. The source
code of our VDIP is available at
https://github.com/Dong-Huo/VDIP-Deconvolution.
| [
{
"version": "v1",
"created": "Tue, 1 Feb 2022 01:33:58 GMT"
},
{
"version": "v2",
"created": "Sat, 6 May 2023 03:11:10 GMT"
},
{
"version": "v3",
"created": "Tue, 6 Jun 2023 02:16:47 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Huo",
"Dong",
""
],
[
"Masoumzadeh",
"Abbas",
""
],
[
"Kushol",
"Rafsanjany",
""
],
[
"Yang",
"Yee-Hong",
""
]
] | TITLE: Blind Image Deconvolution Using Variational Deep Image Prior
ABSTRACT: Conventional deconvolution methods utilize hand-crafted image priors to
constrain the optimization. While deep-learning-based methods have simplified
the optimization by end-to-end training, they fail to generalize well to blurs
unseen in the training dataset. Thus, training image-specific models is
important for higher generalization. Deep image prior (DIP) provides an
approach to optimize the weights of a randomly initialized network with a
single degraded image by maximum a posteriori (MAP), which shows that the
architecture of a network can serve as the hand-crafted image prior. Different
from the conventional hand-crafted image priors that are statistically
obtained, it is hard to find a proper network architecture because the
relationship between images and their corresponding network architectures is
unclear. As a result, the network architecture cannot provide enough constraint
for the latent sharp image. This paper proposes a new variational deep image
prior (VDIP) for blind image deconvolution, which exploits additive
hand-crafted image priors on latent sharp images and approximates a
distribution for each pixel to avoid suboptimal solutions. Our mathematical
analysis shows that the proposed method can better constrain the optimization.
The experimental results further demonstrate that the generated images have
better quality than that of the original DIP on benchmark datasets. The source
code of our VDIP is available at
https://github.com/Dong-Huo/VDIP-Deconvolution.
|
2202.05535 | Kevin Fauvel | Kevin Fauvel, Fuxing Chen, Dario Rossi | A Lightweight, Efficient and Explainable-by-Design Convolutional Neural
Network for Internet Traffic Classification | Accepted for publication in the 29th ACM SIGKDD Conference on
Knowledge Discovery and Data Mining, Long Beach, USA, 2023 | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Traffic classification, i.e. the identification of the type of applications
flowing in a network, is a strategic task for numerous activities (e.g.,
intrusion detection, routing). This task faces some critical challenges that
current deep learning approaches do not address. The design of current
approaches do not take into consideration the fact that networking hardware
(e.g., routers) often runs with limited computational resources. Further, they
do not meet the need for faithful explainability highlighted by regulatory
bodies. Finally, these traffic classifiers are evaluated on small datasets
which fail to reflect the diversity of applications in real-world settings.
Therefore, this paper introduces a new Lightweight, Efficient and
eXplainable-by-design convolutional neural network (LEXNet) for Internet
traffic classification, which relies on a new residual block (for lightweight
and efficiency purposes) and prototype layer (for explainability). Based on a
commercial-grade dataset, our evaluation shows that LEXNet succeeds to maintain
the same accuracy as the best performing state-of-the-art neural network, while
providing the additional features previously mentioned. Moreover, we illustrate
the explainability feature of our approach, which stems from the communication
of detected application prototypes to the end-user, and we highlight the
faithfulness of LEXNet explanations through a comparison with post hoc methods.
| [
{
"version": "v1",
"created": "Fri, 11 Feb 2022 10:21:34 GMT"
},
{
"version": "v2",
"created": "Thu, 4 Aug 2022 16:56:39 GMT"
},
{
"version": "v3",
"created": "Mon, 6 Feb 2023 21:03:04 GMT"
},
{
"version": "v4",
"created": "Mon, 5 Jun 2023 20:52:32 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Fauvel",
"Kevin",
""
],
[
"Chen",
"Fuxing",
""
],
[
"Rossi",
"Dario",
""
]
] | TITLE: A Lightweight, Efficient and Explainable-by-Design Convolutional Neural
Network for Internet Traffic Classification
ABSTRACT: Traffic classification, i.e. the identification of the type of applications
flowing in a network, is a strategic task for numerous activities (e.g.,
intrusion detection, routing). This task faces some critical challenges that
current deep learning approaches do not address. The design of current
approaches do not take into consideration the fact that networking hardware
(e.g., routers) often runs with limited computational resources. Further, they
do not meet the need for faithful explainability highlighted by regulatory
bodies. Finally, these traffic classifiers are evaluated on small datasets
which fail to reflect the diversity of applications in real-world settings.
Therefore, this paper introduces a new Lightweight, Efficient and
eXplainable-by-design convolutional neural network (LEXNet) for Internet
traffic classification, which relies on a new residual block (for lightweight
and efficiency purposes) and prototype layer (for explainability). Based on a
commercial-grade dataset, our evaluation shows that LEXNet succeeds to maintain
the same accuracy as the best performing state-of-the-art neural network, while
providing the additional features previously mentioned. Moreover, we illustrate
the explainability feature of our approach, which stems from the communication
of detected application prototypes to the end-user, and we highlight the
faithfulness of LEXNet explanations through a comparison with post hoc methods.
|
2204.03939 | Rong Ye | Rong Ye, Chengqi Zhao, Tom Ko, Chutong Meng, Tao Wang, Mingxuan Wang,
Jun Cao | GigaST: A 10,000-hour Pseudo Speech Translation Corpus | Accepted at Interspeech 2023. GigaST dataset is available at
https://st-benchmark.github.io/resources/GigaST | null | null | null | cs.CL cs.SD eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper introduces GigaST, a large-scale pseudo speech translation (ST)
corpus. We create the corpus by translating the text in GigaSpeech, an English
ASR corpus, into German and Chinese. The training set is translated by a strong
machine translation system and the test set is translated by human. ST models
trained with an addition of our corpus obtain new state-of-the-art results on
the MuST-C English-German benchmark test set. We provide a detailed description
of the translation process and verify its quality. We make the translated text
data public and hope to facilitate research in speech translation.
Additionally, we also release the training scripts on NeurST to make it easy to
replicate our systems. GigaST dataset is available at
https://st-benchmark.github.io/resources/GigaST.
| [
{
"version": "v1",
"created": "Fri, 8 Apr 2022 08:59:33 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 12:48:48 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Ye",
"Rong",
""
],
[
"Zhao",
"Chengqi",
""
],
[
"Ko",
"Tom",
""
],
[
"Meng",
"Chutong",
""
],
[
"Wang",
"Tao",
""
],
[
"Wang",
"Mingxuan",
""
],
[
"Cao",
"Jun",
""
]
] | TITLE: GigaST: A 10,000-hour Pseudo Speech Translation Corpus
ABSTRACT: This paper introduces GigaST, a large-scale pseudo speech translation (ST)
corpus. We create the corpus by translating the text in GigaSpeech, an English
ASR corpus, into German and Chinese. The training set is translated by a strong
machine translation system and the test set is translated by human. ST models
trained with an addition of our corpus obtain new state-of-the-art results on
the MuST-C English-German benchmark test set. We provide a detailed description
of the translation process and verify its quality. We make the translated text
data public and hope to facilitate research in speech translation.
Additionally, we also release the training scripts on NeurST to make it easy to
replicate our systems. GigaST dataset is available at
https://st-benchmark.github.io/resources/GigaST.
|
2204.06638 | Daniel Liang | Daniel Liang | Clifford Circuits can be Properly PAC Learned if and only if
$\textsf{RP}=\textsf{NP}$ | 30 pages, 2 figures | Quantum 7, 1036 (2023) | 10.22331/q-2023-06-07-1036 | null | quant-ph cs.CC cs.LG | http://creativecommons.org/licenses/by/4.0/ | Given a dataset of input states, measurements, and probabilities, is it
possible to efficiently predict the measurement probabilities associated with a
quantum circuit? Recent work of Caro and Datta (2020) studied the problem of
PAC learning quantum circuits in an information theoretic sense, leaving open
questions of computational efficiency. In particular, one candidate class of
circuits for which an efficient learner might have been possible was that of
Clifford circuits, since the corresponding set of states generated by such
circuits, called stabilizer states, are known to be efficiently PAC learnable
(Rocchetto 2018). Here we provide a negative result, showing that proper
learning of CNOT circuits is hard for classical learners unless $\textsf{RP} =
\textsf{NP}$. As the classical analogue and subset of Clifford circuits, this
naturally leads to a hardness result for Clifford circuits as well.
Additionally, we show that if $\textsf{RP} = \textsf{NP}$ then there would
exist efficient proper learning algorithms for CNOT and Clifford circuits. By
similar arguments, we also find that an efficient proper quantum learner for
such circuits exists if and only if $\textsf{NP} \subseteq \textsf{RQP}$.
| [
{
"version": "v1",
"created": "Wed, 13 Apr 2022 21:07:53 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Apr 2022 15:55:17 GMT"
},
{
"version": "v3",
"created": "Wed, 11 May 2022 16:02:29 GMT"
},
{
"version": "v4",
"created": "Tue, 6 Jun 2023 00:15:47 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Liang",
"Daniel",
""
]
] | TITLE: Clifford Circuits can be Properly PAC Learned if and only if
$\textsf{RP}=\textsf{NP}$
ABSTRACT: Given a dataset of input states, measurements, and probabilities, is it
possible to efficiently predict the measurement probabilities associated with a
quantum circuit? Recent work of Caro and Datta (2020) studied the problem of
PAC learning quantum circuits in an information theoretic sense, leaving open
questions of computational efficiency. In particular, one candidate class of
circuits for which an efficient learner might have been possible was that of
Clifford circuits, since the corresponding set of states generated by such
circuits, called stabilizer states, are known to be efficiently PAC learnable
(Rocchetto 2018). Here we provide a negative result, showing that proper
learning of CNOT circuits is hard for classical learners unless $\textsf{RP} =
\textsf{NP}$. As the classical analogue and subset of Clifford circuits, this
naturally leads to a hardness result for Clifford circuits as well.
Additionally, we show that if $\textsf{RP} = \textsf{NP}$ then there would
exist efficient proper learning algorithms for CNOT and Clifford circuits. By
similar arguments, we also find that an efficient proper quantum learner for
such circuits exists if and only if $\textsf{NP} \subseteq \textsf{RQP}$.
|
2206.04285 | Mehrdad Khatir | Mehrdad Khatir, Nurendra Choudhary, Sutanay Choudhury, Khushbu
Agarwal, Chandan K. Reddy | A Unification Framework for Euclidean and Hyperbolic Graph Neural
Networks | null | null | null | null | cs.LG cs.NE | http://creativecommons.org/licenses/by/4.0/ | Hyperbolic neural networks can effectively capture the inherent hierarchy of
graph datasets, and consequently a powerful choice of GNNs. However, they
entangle multiple incongruent (gyro-)vector spaces within a layer, which makes
them limited in terms of generalization and scalability. In this work, we
propose the Poincare disk model as our search space, and apply all
approximations on the disk (as if the disk is a tangent space derived from the
origin), thus getting rid of all inter-space transformations. Such an approach
enables us to propose a hyperbolic normalization layer and to further simplify
the entire hyperbolic model to a Euclidean model cascaded with our hyperbolic
normalization layer. We applied our proposed nonlinear hyperbolic normalization
to the current state-of-the-art homogeneous and multi-relational graph
networks. We demonstrate that our model not only leverages the power of
Euclidean networks such as interpretability and efficient execution of various
model components, but also outperforms both Euclidean and hyperbolic
counterparts on various benchmarks. Our code is made publicly available at
https://github.com/oom-debugger/ijcai23.
| [
{
"version": "v1",
"created": "Thu, 9 Jun 2022 05:33:02 GMT"
},
{
"version": "v2",
"created": "Sat, 21 Jan 2023 17:53:13 GMT"
},
{
"version": "v3",
"created": "Tue, 6 Jun 2023 12:38:45 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Khatir",
"Mehrdad",
""
],
[
"Choudhary",
"Nurendra",
""
],
[
"Choudhury",
"Sutanay",
""
],
[
"Agarwal",
"Khushbu",
""
],
[
"Reddy",
"Chandan K.",
""
]
] | TITLE: A Unification Framework for Euclidean and Hyperbolic Graph Neural
Networks
ABSTRACT: Hyperbolic neural networks can effectively capture the inherent hierarchy of
graph datasets, and consequently a powerful choice of GNNs. However, they
entangle multiple incongruent (gyro-)vector spaces within a layer, which makes
them limited in terms of generalization and scalability. In this work, we
propose the Poincare disk model as our search space, and apply all
approximations on the disk (as if the disk is a tangent space derived from the
origin), thus getting rid of all inter-space transformations. Such an approach
enables us to propose a hyperbolic normalization layer and to further simplify
the entire hyperbolic model to a Euclidean model cascaded with our hyperbolic
normalization layer. We applied our proposed nonlinear hyperbolic normalization
to the current state-of-the-art homogeneous and multi-relational graph
networks. We demonstrate that our model not only leverages the power of
Euclidean networks such as interpretability and efficient execution of various
model components, but also outperforms both Euclidean and hyperbolic
counterparts on various benchmarks. Our code is made publicly available at
https://github.com/oom-debugger/ijcai23.
|
2206.04882 | Ziqi Chen | Ziqi Chen, Oluwatosin R. Ayinde, James R. Fuchs, Huan Sun, Xia Ning | $\mathsf{G^2Retro}$ as a Two-Step Graph Generative Models for
Retrosynthesis Prediction | null | Commun Chem 6, 102 (2023) | 10.1038/s42004-023-00897-3 | null | cs.LG physics.chem-ph q-bio.BM | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Retrosynthesis is a procedure where a target molecule is transformed into
potential reactants and thus the synthesis routes can be identified. Recently,
computational approaches have been developed to accelerate the design of
synthesis routes. In this paper, we develop a generative framework
$\mathsf{G^2Retro}$ for one-step retrosynthesis prediction. $\mathsf{G^2Retro}$
imitates the reversed logic of synthetic reactions. It first predicts the
reaction centers in the target molecules (products), identifies the synthons
needed to assemble the products, and transforms these synthons into reactants.
$\mathsf{G^2Retro}$ defines a comprehensive set of reaction center types, and
learns from the molecular graphs of the products to predict potential reaction
centers. To complete synthons into reactants, $\mathsf{G^2Retro}$ considers all
the involved synthon structures and the product structures to identify the
optimal completion paths, and accordingly attaches small substructures
sequentially to the synthons. Here we show that $\mathsf{G^2Retro}$ is able to
better predict the reactants for given products in the benchmark dataset than
the state-of-the-art methods.
| [
{
"version": "v1",
"created": "Fri, 10 Jun 2022 05:34:12 GMT"
},
{
"version": "v2",
"created": "Sun, 20 Nov 2022 18:32:43 GMT"
},
{
"version": "v3",
"created": "Mon, 5 Jun 2023 20:58:47 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Chen",
"Ziqi",
""
],
[
"Ayinde",
"Oluwatosin R.",
""
],
[
"Fuchs",
"James R.",
""
],
[
"Sun",
"Huan",
""
],
[
"Ning",
"Xia",
""
]
] | TITLE: $\mathsf{G^2Retro}$ as a Two-Step Graph Generative Models for
Retrosynthesis Prediction
ABSTRACT: Retrosynthesis is a procedure where a target molecule is transformed into
potential reactants and thus the synthesis routes can be identified. Recently,
computational approaches have been developed to accelerate the design of
synthesis routes. In this paper, we develop a generative framework
$\mathsf{G^2Retro}$ for one-step retrosynthesis prediction. $\mathsf{G^2Retro}$
imitates the reversed logic of synthetic reactions. It first predicts the
reaction centers in the target molecules (products), identifies the synthons
needed to assemble the products, and transforms these synthons into reactants.
$\mathsf{G^2Retro}$ defines a comprehensive set of reaction center types, and
learns from the molecular graphs of the products to predict potential reaction
centers. To complete synthons into reactants, $\mathsf{G^2Retro}$ considers all
the involved synthon structures and the product structures to identify the
optimal completion paths, and accordingly attaches small substructures
sequentially to the synthons. Here we show that $\mathsf{G^2Retro}$ is able to
better predict the reactants for given products in the benchmark dataset than
the state-of-the-art methods.
|
2206.05731 | Zongyuan Huang | Zongyuan Huang, Shengyuan Xu, Menghan Wang, Hansi Wu, Yanyan Xu,
Yaohui Jin | Human Mobility Prediction with Causal and Spatial-constrained Multi-task
Network | Updated version (We have corrected the title error in the previous
version.) | null | null | null | cs.AI cs.IR | http://creativecommons.org/licenses/by/4.0/ | Modeling human mobility helps to understand how people are accessing
resources and physically contacting with each other in cities, and thus
contributes to various applications such as urban planning, epidemic control,
and location-based advertisement. Next location prediction is one decisive task
in individual human mobility modeling and is usually viewed as sequence
modeling, solved with Markov or RNN-based methods. However, the existing models
paid little attention to the logic of individual travel decisions and the
reproducibility of the collective behavior of population. To this end, we
propose a Causal and Spatial-constrained Long and Short-term Learner (CSLSL)
for next location prediction. CSLSL utilizes a causal structure based on
multi-task learning to explicitly model the
"\textit{when$\rightarrow$what$\rightarrow$where}", a.k.a.
"\textit{time$\rightarrow$activity$\rightarrow$location}" decision logic. We
next propose a spatial-constrained loss function as an auxiliary task, to
ensure the consistency between the predicted and actual spatial distribution of
travelers' destinations. Moreover, CSLSL adopts modules named Long and
Short-term Capturer (LSC) to learn the transition regularities across different
time spans. Extensive experiments on three real-world datasets show promising
performance improvements of CSLSL over baselines and confirm the effectiveness
of introducing the causality and consistency constraints. The implementation is
available at https://github.com/urbanmobility/CSLSL.
| [
{
"version": "v1",
"created": "Sun, 12 Jun 2022 13:03:47 GMT"
},
{
"version": "v2",
"created": "Sun, 17 Jul 2022 15:59:33 GMT"
},
{
"version": "v3",
"created": "Wed, 1 Feb 2023 07:47:28 GMT"
},
{
"version": "v4",
"created": "Tue, 6 Jun 2023 03:34:34 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Huang",
"Zongyuan",
""
],
[
"Xu",
"Shengyuan",
""
],
[
"Wang",
"Menghan",
""
],
[
"Wu",
"Hansi",
""
],
[
"Xu",
"Yanyan",
""
],
[
"Jin",
"Yaohui",
""
]
] | TITLE: Human Mobility Prediction with Causal and Spatial-constrained Multi-task
Network
ABSTRACT: Modeling human mobility helps to understand how people are accessing
resources and physically contacting with each other in cities, and thus
contributes to various applications such as urban planning, epidemic control,
and location-based advertisement. Next location prediction is one decisive task
in individual human mobility modeling and is usually viewed as sequence
modeling, solved with Markov or RNN-based methods. However, the existing models
paid little attention to the logic of individual travel decisions and the
reproducibility of the collective behavior of population. To this end, we
propose a Causal and Spatial-constrained Long and Short-term Learner (CSLSL)
for next location prediction. CSLSL utilizes a causal structure based on
multi-task learning to explicitly model the
"\textit{when$\rightarrow$what$\rightarrow$where}", a.k.a.
"\textit{time$\rightarrow$activity$\rightarrow$location}" decision logic. We
next propose a spatial-constrained loss function as an auxiliary task, to
ensure the consistency between the predicted and actual spatial distribution of
travelers' destinations. Moreover, CSLSL adopts modules named Long and
Short-term Capturer (LSC) to learn the transition regularities across different
time spans. Extensive experiments on three real-world datasets show promising
performance improvements of CSLSL over baselines and confirm the effectiveness
of introducing the causality and consistency constraints. The implementation is
available at https://github.com/urbanmobility/CSLSL.
|
2206.09959 | Ali Hatamizadeh | Ali Hatamizadeh, Hongxu Yin, Greg Heinrich, Jan Kautz, and Pavlo
Molchanov | Global Context Vision Transformers | Accepted to ICML 2023 | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We propose global context vision transformer (GC ViT), a novel architecture
that enhances parameter and compute utilization for computer vision. Our method
leverages global context self-attention modules, joint with standard local
self-attention, to effectively and efficiently model both long and short-range
spatial interactions, without the need for expensive operations such as
computing attention masks or shifting local windows. In addition, we address
the lack of the inductive bias in ViTs, and propose to leverage a modified
fused inverted residual blocks in our architecture. Our proposed GC ViT
achieves state-of-the-art results across image classification, object detection
and semantic segmentation tasks. On ImageNet-1K dataset for classification, the
variants of GC ViT with 51M, 90M and 201M parameters achieve 84.3%, 85.0% and
85.7% Top-1 accuracy, respectively, at 224 image resolution and without any
pre-training, hence surpassing comparably-sized prior art such as CNN-based
ConvNeXt and ViT-based MaxViT and Swin Transformer by a large margin.
Pre-trained GC ViT backbones in downstream tasks of object detection, instance
segmentation, and semantic segmentation using MS COCO and ADE20K datasets
outperform prior work consistently. Specifically, GC ViT with a 4-scale DINO
detection head achieves a box AP of 58.3 on MS COCO dataset.
| [
{
"version": "v1",
"created": "Mon, 20 Jun 2022 18:42:44 GMT"
},
{
"version": "v2",
"created": "Wed, 7 Sep 2022 21:02:00 GMT"
},
{
"version": "v3",
"created": "Sat, 1 Oct 2022 03:40:57 GMT"
},
{
"version": "v4",
"created": "Mon, 6 Feb 2023 04:38:57 GMT"
},
{
"version": "v5",
"created": "Tue, 6 Jun 2023 08:17:18 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Hatamizadeh",
"Ali",
""
],
[
"Yin",
"Hongxu",
""
],
[
"Heinrich",
"Greg",
""
],
[
"Kautz",
"Jan",
""
],
[
"Molchanov",
"Pavlo",
""
]
] | TITLE: Global Context Vision Transformers
ABSTRACT: We propose global context vision transformer (GC ViT), a novel architecture
that enhances parameter and compute utilization for computer vision. Our method
leverages global context self-attention modules, joint with standard local
self-attention, to effectively and efficiently model both long and short-range
spatial interactions, without the need for expensive operations such as
computing attention masks or shifting local windows. In addition, we address
the lack of the inductive bias in ViTs, and propose to leverage a modified
fused inverted residual blocks in our architecture. Our proposed GC ViT
achieves state-of-the-art results across image classification, object detection
and semantic segmentation tasks. On ImageNet-1K dataset for classification, the
variants of GC ViT with 51M, 90M and 201M parameters achieve 84.3%, 85.0% and
85.7% Top-1 accuracy, respectively, at 224 image resolution and without any
pre-training, hence surpassing comparably-sized prior art such as CNN-based
ConvNeXt and ViT-based MaxViT and Swin Transformer by a large margin.
Pre-trained GC ViT backbones in downstream tasks of object detection, instance
segmentation, and semantic segmentation using MS COCO and ADE20K datasets
outperform prior work consistently. Specifically, GC ViT with a 4-scale DINO
detection head achieves a box AP of 58.3 on MS COCO dataset.
|
2206.10713 | Rudrajit Das | Rudrajit Das, Satyen Kale, Zheng Xu, Tong Zhang, Sujay Sanghavi | Beyond Uniform Lipschitz Condition in Differentially Private
Optimization | To appear in ICML 2023 | null | null | null | cs.LG stat.ML | http://creativecommons.org/licenses/by/4.0/ | Most prior results on differentially private stochastic gradient descent
(DP-SGD) are derived under the simplistic assumption of uniform Lipschitzness,
i.e., the per-sample gradients are uniformly bounded. We generalize uniform
Lipschitzness by assuming that the per-sample gradients have sample-dependent
upper bounds, i.e., per-sample Lipschitz constants, which themselves may be
unbounded. We provide principled guidance on choosing the clip norm in DP-SGD
for convex over-parameterized settings satisfying our general version of
Lipschitzness when the per-sample Lipschitz constants are bounded;
specifically, we recommend tuning the clip norm only till values up to the
minimum per-sample Lipschitz constant. This finds application in the private
training of a softmax layer on top of a deep network pre-trained on public
data. We verify the efficacy of our recommendation via experiments on 8
datasets. Furthermore, we provide new convergence results for DP-SGD on convex
and nonconvex functions when the Lipschitz constants are unbounded but have
bounded moments, i.e., they are heavy-tailed.
| [
{
"version": "v1",
"created": "Tue, 21 Jun 2022 20:11:30 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 01:26:35 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Das",
"Rudrajit",
""
],
[
"Kale",
"Satyen",
""
],
[
"Xu",
"Zheng",
""
],
[
"Zhang",
"Tong",
""
],
[
"Sanghavi",
"Sujay",
""
]
] | TITLE: Beyond Uniform Lipschitz Condition in Differentially Private
Optimization
ABSTRACT: Most prior results on differentially private stochastic gradient descent
(DP-SGD) are derived under the simplistic assumption of uniform Lipschitzness,
i.e., the per-sample gradients are uniformly bounded. We generalize uniform
Lipschitzness by assuming that the per-sample gradients have sample-dependent
upper bounds, i.e., per-sample Lipschitz constants, which themselves may be
unbounded. We provide principled guidance on choosing the clip norm in DP-SGD
for convex over-parameterized settings satisfying our general version of
Lipschitzness when the per-sample Lipschitz constants are bounded;
specifically, we recommend tuning the clip norm only till values up to the
minimum per-sample Lipschitz constant. This finds application in the private
training of a softmax layer on top of a deep network pre-trained on public
data. We verify the efficacy of our recommendation via experiments on 8
datasets. Furthermore, we provide new convergence results for DP-SGD on convex
and nonconvex functions when the Lipschitz constants are unbounded but have
bounded moments, i.e., they are heavy-tailed.
|
2206.10858 | Changming Xu | Changming Xu, Gagandeep Singh | Robust Universal Adversarial Perturbations | 16 pages, 3 figures | null | null | null | cs.LG cs.CR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Universal Adversarial Perturbations (UAPs) are imperceptible, image-agnostic
vectors that cause deep neural networks (DNNs) to misclassify inputs with high
probability. In practical attack scenarios, adversarial perturbations may
undergo transformations such as changes in pixel intensity, scaling, etc.
before being added to DNN inputs. Existing methods do not create UAPs robust to
these real-world transformations, thereby limiting their applicability in
practical attack scenarios. In this work, we introduce and formulate UAPs
robust against real-world transformations. We build an iterative algorithm
using probabilistic robustness bounds and construct such UAPs robust to
transformations generated by composing arbitrary sub-differentiable
transformation functions. We perform an extensive evaluation on the popular
CIFAR-10 and ILSVRC 2012 datasets measuring our UAPs' robustness under a wide
range common, real-world transformations such as rotation, contrast changes,
etc. We further show that by using a set of primitive transformations our
method can generalize well to unseen transformations such as fog, JPEG
compression, etc. Our results show that our method can generate UAPs up to 23%
more robust than state-of-the-art baselines.
| [
{
"version": "v1",
"created": "Wed, 22 Jun 2022 06:05:30 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 05:16:38 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Xu",
"Changming",
""
],
[
"Singh",
"Gagandeep",
""
]
] | TITLE: Robust Universal Adversarial Perturbations
ABSTRACT: Universal Adversarial Perturbations (UAPs) are imperceptible, image-agnostic
vectors that cause deep neural networks (DNNs) to misclassify inputs with high
probability. In practical attack scenarios, adversarial perturbations may
undergo transformations such as changes in pixel intensity, scaling, etc.
before being added to DNN inputs. Existing methods do not create UAPs robust to
these real-world transformations, thereby limiting their applicability in
practical attack scenarios. In this work, we introduce and formulate UAPs
robust against real-world transformations. We build an iterative algorithm
using probabilistic robustness bounds and construct such UAPs robust to
transformations generated by composing arbitrary sub-differentiable
transformation functions. We perform an extensive evaluation on the popular
CIFAR-10 and ILSVRC 2012 datasets measuring our UAPs' robustness under a wide
range common, real-world transformations such as rotation, contrast changes,
etc. We further show that by using a set of primitive transformations our
method can generalize well to unseen transformations such as fog, JPEG
compression, etc. Our results show that our method can generate UAPs up to 23%
more robust than state-of-the-art baselines.
|
2207.02715 | Niklas Kochdumper Dr. | Niklas Kochdumper, Christian Schilling, Matthias Althoff, Stanley Bak | Open- and Closed-Loop Neural Network Verification using Polynomial
Zonotopes | null | NFM 2023 | 10.1007/978-3-031-33170-1_2 | null | cs.CV cs.AI cs.LO cs.SY eess.SY | http://creativecommons.org/licenses/by/4.0/ | We present a novel approach to efficiently compute tight non-convex
enclosures of the image through neural networks with ReLU, sigmoid, or
hyperbolic tangent activation functions. In particular, we abstract the
input-output relation of each neuron by a polynomial approximation, which is
evaluated in a set-based manner using polynomial zonotopes. While our approach
can also can be beneficial for open-loop neural network verification, our main
application is reachability analysis of neural network controlled systems,
where polynomial zonotopes are able to capture the non-convexity caused by the
neural network as well as the system dynamics. This results in a superior
performance compared to other methods, as we demonstrate on various benchmarks.
| [
{
"version": "v1",
"created": "Wed, 6 Jul 2022 14:39:19 GMT"
},
{
"version": "v2",
"created": "Tue, 18 Apr 2023 02:58:42 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Kochdumper",
"Niklas",
""
],
[
"Schilling",
"Christian",
""
],
[
"Althoff",
"Matthias",
""
],
[
"Bak",
"Stanley",
""
]
] | TITLE: Open- and Closed-Loop Neural Network Verification using Polynomial
Zonotopes
ABSTRACT: We present a novel approach to efficiently compute tight non-convex
enclosures of the image through neural networks with ReLU, sigmoid, or
hyperbolic tangent activation functions. In particular, we abstract the
input-output relation of each neuron by a polynomial approximation, which is
evaluated in a set-based manner using polynomial zonotopes. While our approach
can also can be beneficial for open-loop neural network verification, our main
application is reachability analysis of neural network controlled systems,
where polynomial zonotopes are able to capture the non-convexity caused by the
neural network as well as the system dynamics. This results in a superior
performance compared to other methods, as we demonstrate on various benchmarks.
|
2209.05098 | David Erzmann | S\"oren Dittmer, David Erzmann, Henrik Harms, Peter Maass | SELTO: Sample-Efficient Learned Topology Optimization | 25 pages, 10 figures, submitted to the International Journal for
Numerical Methods in Engineering | null | null | null | cs.LG cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent developments in Deep Learning (DL) suggest a vast potential for
Topology Optimization (TO). However, while there are some promising attempts,
the subfield still lacks a firm footing regarding basic methods and datasets.
We aim to address both points. First, we explore physics-based preprocessing
and equivariant networks to create sample-efficient components for TO DL
pipelines. We evaluate them in a large-scale ablation study using end-to-end
supervised training. The results demonstrate a drastic improvement in sample
efficiency and the predictions' physical correctness. Second, to improve
comparability and future progress, we publish the two first TO datasets
containing problems and corresponding ground truth solutions.
| [
{
"version": "v1",
"created": "Mon, 12 Sep 2022 09:02:00 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 11:02:06 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Dittmer",
"Sören",
""
],
[
"Erzmann",
"David",
""
],
[
"Harms",
"Henrik",
""
],
[
"Maass",
"Peter",
""
]
] | TITLE: SELTO: Sample-Efficient Learned Topology Optimization
ABSTRACT: Recent developments in Deep Learning (DL) suggest a vast potential for
Topology Optimization (TO). However, while there are some promising attempts,
the subfield still lacks a firm footing regarding basic methods and datasets.
We aim to address both points. First, we explore physics-based preprocessing
and equivariant networks to create sample-efficient components for TO DL
pipelines. We evaluate them in a large-scale ablation study using end-to-end
supervised training. The results demonstrate a drastic improvement in sample
efficiency and the predictions' physical correctness. Second, to improve
comparability and future progress, we publish the two first TO datasets
containing problems and corresponding ground truth solutions.
|
2209.08284 | Yujie Lu | Yujie Lu, Siqi Ouyang, Kairui Zhou | Structured Knowledge Grounding for Question Answering | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Can language models (LM) ground question-answering (QA) tasks in the
knowledge base via inherent relational reasoning ability? While previous models
that use only LMs have seen some success on many QA tasks, more recent methods
include knowledge graphs (KG) to complement LMs with their more logic-driven
implicit knowledge. However, effectively extracting information from structured
data, like KGs, empowers LMs to remain an open question, and current models
rely on graph techniques to extract knowledge. In this paper, we propose to
solely leverage the LMs to combine the language and knowledge for knowledge
based question-answering with flexibility, breadth of coverage and structured
reasoning. Specifically, we devise a knowledge construction method that
retrieves the relevant context with a dynamic hop, which expresses more
comprehensivenes than traditional GNN-based techniques. And we devise a deep
fusion mechanism to further bridge the information exchanging bottleneck
between the language and the knowledge. Extensive experiments show that our
model consistently demonstrates its state-of-the-art performance over
CommensenseQA benchmark, showcasing the possibility to leverage LMs solely to
robustly ground QA into the knowledge base.
| [
{
"version": "v1",
"created": "Sat, 17 Sep 2022 08:48:50 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Oct 2022 01:25:47 GMT"
},
{
"version": "v3",
"created": "Mon, 5 Jun 2023 20:41:18 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Lu",
"Yujie",
""
],
[
"Ouyang",
"Siqi",
""
],
[
"Zhou",
"Kairui",
""
]
] | TITLE: Structured Knowledge Grounding for Question Answering
ABSTRACT: Can language models (LM) ground question-answering (QA) tasks in the
knowledge base via inherent relational reasoning ability? While previous models
that use only LMs have seen some success on many QA tasks, more recent methods
include knowledge graphs (KG) to complement LMs with their more logic-driven
implicit knowledge. However, effectively extracting information from structured
data, like KGs, empowers LMs to remain an open question, and current models
rely on graph techniques to extract knowledge. In this paper, we propose to
solely leverage the LMs to combine the language and knowledge for knowledge
based question-answering with flexibility, breadth of coverage and structured
reasoning. Specifically, we devise a knowledge construction method that
retrieves the relevant context with a dynamic hop, which expresses more
comprehensivenes than traditional GNN-based techniques. And we devise a deep
fusion mechanism to further bridge the information exchanging bottleneck
between the language and the knowledge. Extensive experiments show that our
model consistently demonstrates its state-of-the-art performance over
CommensenseQA benchmark, showcasing the possibility to leverage LMs solely to
robustly ground QA into the knowledge base.
|
2209.15245 | Jianyi Zhang | Jianyi Zhang, Ang Li, Minxue Tang, Jingwei Sun, Xiang Chen, Fan Zhang,
Changyou Chen, Yiran Chen, Hai Li | Fed-CBS: A Heterogeneity-Aware Client Sampling Mechanism for Federated
Learning via Class-Imbalance Reduction | ICML 2023 | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Due to limited communication capacities of edge devices, most existing
federated learning (FL) methods randomly select only a subset of devices to
participate in training for each communication round. Compared with engaging
all the available clients, the random-selection mechanism can lead to
significant performance degradation on non-IID (independent and identically
distributed) data. In this paper, we show our key observation that the
essential reason resulting in such performance degradation is the
class-imbalance of the grouped data from randomly selected clients. Based on
our key observation, we design an efficient heterogeneity-aware client sampling
mechanism, i.e., Federated Class-balanced Sampling (Fed-CBS), which can
effectively reduce class-imbalance of the group dataset from the intentionally
selected clients. In particular, we propose a measure of class-imbalance and
then employ homomorphic encryption to derive this measure in a
privacy-preserving way. Based on this measure, we also design a
computation-efficient client sampling strategy, such that the actively selected
clients will generate a more class-balanced grouped dataset with theoretical
guarantees. Extensive experimental results demonstrate Fed-CBS outperforms the
status quo approaches. Furthermore, it achieves comparable or even better
performance than the ideal setting where all the available clients participate
in the FL training.
| [
{
"version": "v1",
"created": "Fri, 30 Sep 2022 05:42:56 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 06:27:08 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Zhang",
"Jianyi",
""
],
[
"Li",
"Ang",
""
],
[
"Tang",
"Minxue",
""
],
[
"Sun",
"Jingwei",
""
],
[
"Chen",
"Xiang",
""
],
[
"Zhang",
"Fan",
""
],
[
"Chen",
"Changyou",
""
],
[
"Chen",
"Yiran",
""
],
[
"Li",
"Hai",
""
]
] | TITLE: Fed-CBS: A Heterogeneity-Aware Client Sampling Mechanism for Federated
Learning via Class-Imbalance Reduction
ABSTRACT: Due to limited communication capacities of edge devices, most existing
federated learning (FL) methods randomly select only a subset of devices to
participate in training for each communication round. Compared with engaging
all the available clients, the random-selection mechanism can lead to
significant performance degradation on non-IID (independent and identically
distributed) data. In this paper, we show our key observation that the
essential reason resulting in such performance degradation is the
class-imbalance of the grouped data from randomly selected clients. Based on
our key observation, we design an efficient heterogeneity-aware client sampling
mechanism, i.e., Federated Class-balanced Sampling (Fed-CBS), which can
effectively reduce class-imbalance of the group dataset from the intentionally
selected clients. In particular, we propose a measure of class-imbalance and
then employ homomorphic encryption to derive this measure in a
privacy-preserving way. Based on this measure, we also design a
computation-efficient client sampling strategy, such that the actively selected
clients will generate a more class-balanced grouped dataset with theoretical
guarantees. Extensive experimental results demonstrate Fed-CBS outperforms the
status quo approaches. Furthermore, it achieves comparable or even better
performance than the ideal setting where all the available clients participate
in the FL training.
|
2210.00586 | Ali Borji | Ali Borji | Generated Faces in the Wild: Quantitative Comparison of Stable
Diffusion, Midjourney and DALL-E 2 | dataset link udated! | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | The field of image synthesis has made great strides in the last couple of
years. Recent models are capable of generating images with astonishing quality.
Fine-grained evaluation of these models on some interesting categories such as
faces is still missing. Here, we conduct a quantitative comparison of three
popular systems including Stable Diffusion, Midjourney, and DALL-E 2 in their
ability to generate photorealistic faces in the wild. We find that Stable
Diffusion generates better faces than the other systems, according to the FID
score. We also introduce a dataset of generated faces in the wild dubbed GFW,
including a total of 15,076 faces. Furthermore, we hope that our study spurs
follow-up research in assessing the generative models and improving them. Data
and code are available at data and code, respectively.
| [
{
"version": "v1",
"created": "Sun, 2 Oct 2022 17:53:08 GMT"
},
{
"version": "v2",
"created": "Mon, 5 Jun 2023 20:25:22 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Borji",
"Ali",
""
]
] | TITLE: Generated Faces in the Wild: Quantitative Comparison of Stable
Diffusion, Midjourney and DALL-E 2
ABSTRACT: The field of image synthesis has made great strides in the last couple of
years. Recent models are capable of generating images with astonishing quality.
Fine-grained evaluation of these models on some interesting categories such as
faces is still missing. Here, we conduct a quantitative comparison of three
popular systems including Stable Diffusion, Midjourney, and DALL-E 2 in their
ability to generate photorealistic faces in the wild. We find that Stable
Diffusion generates better faces than the other systems, according to the FID
score. We also introduce a dataset of generated faces in the wild dubbed GFW,
including a total of 15,076 faces. Furthermore, we hope that our study spurs
follow-up research in assessing the generative models and improving them. Data
and code are available at data and code, respectively.
|
2210.02969 | Seonghyeon Ye | Seonghyeon Ye, Doyoung Kim, Joel Jang, Joongbo Shin, Minjoon Seo | Guess the Instruction! Flipped Learning Makes Language Models Stronger
Zero-Shot Learners | ICLR 2023 | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Meta-training, which fine-tunes the language model (LM) on various downstream
tasks by maximizing the likelihood of the target label given the task
instruction and input instance, has improved the zero-shot task generalization
performance. However, meta-trained LMs still struggle to generalize to
challenging tasks containing novel labels unseen during meta-training. In this
paper, we propose Flipped Learning, an alternative method of meta-training
which trains the LM to generate the task instruction given the input instance
and label. During inference, the LM trained with Flipped Learning, referred to
as Flipped, selects the label option that is most likely to generate the task
instruction. On 14 tasks of the BIG-bench benchmark, the 11B-sized Flipped
outperforms zero-shot T0-11B and even a 16 times larger 3-shot GPT-3 (175B) on
average by 8.4% and 9.7% points, respectively. Flipped gives particularly large
improvements on tasks with unseen labels, outperforming T0-11B by up to +20%
average F1 score. This indicates that the strong task generalization of Flipped
comes from improved generalization to novel labels. We release our code at
https://github.com/seonghyeonye/Flipped-Learning.
| [
{
"version": "v1",
"created": "Thu, 6 Oct 2022 15:00:47 GMT"
},
{
"version": "v2",
"created": "Tue, 11 Oct 2022 14:42:48 GMT"
},
{
"version": "v3",
"created": "Sun, 4 Dec 2022 16:36:28 GMT"
},
{
"version": "v4",
"created": "Tue, 6 Jun 2023 11:03:31 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Ye",
"Seonghyeon",
""
],
[
"Kim",
"Doyoung",
""
],
[
"Jang",
"Joel",
""
],
[
"Shin",
"Joongbo",
""
],
[
"Seo",
"Minjoon",
""
]
] | TITLE: Guess the Instruction! Flipped Learning Makes Language Models Stronger
Zero-Shot Learners
ABSTRACT: Meta-training, which fine-tunes the language model (LM) on various downstream
tasks by maximizing the likelihood of the target label given the task
instruction and input instance, has improved the zero-shot task generalization
performance. However, meta-trained LMs still struggle to generalize to
challenging tasks containing novel labels unseen during meta-training. In this
paper, we propose Flipped Learning, an alternative method of meta-training
which trains the LM to generate the task instruction given the input instance
and label. During inference, the LM trained with Flipped Learning, referred to
as Flipped, selects the label option that is most likely to generate the task
instruction. On 14 tasks of the BIG-bench benchmark, the 11B-sized Flipped
outperforms zero-shot T0-11B and even a 16 times larger 3-shot GPT-3 (175B) on
average by 8.4% and 9.7% points, respectively. Flipped gives particularly large
improvements on tasks with unseen labels, outperforming T0-11B by up to +20%
average F1 score. This indicates that the strong task generalization of Flipped
comes from improved generalization to novel labels. We release our code at
https://github.com/seonghyeonye/Flipped-Learning.
|
2210.04185 | Zekun Li | Zekun Li, Wenhu Chen, Shiyang Li, Hong Wang, Jing Qian, Xifeng Yan | Controllable Dialogue Simulation with In-Context Learning | EMNLP 2022 Findings, code and data are available at
https://github.com/Leezekun/dialogic | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Building dialogue systems requires a large corpus of annotated dialogues.
Such datasets are usually created via crowdsourcing, which is expensive and
time-consuming. In this paper, we propose \textsc{Dialogic}, a novel dialogue
simulation method based on large language model in-context learning to automate
dataset creation. Seeded with a few annotated dialogues, \textsc{Dialogic}
automatically selects in-context examples for demonstration and prompts GPT-3
to generate new dialogues and annotations in a controllable way. Our method can
rapidly expand a small set of dialogue data with minimum or zero \textit{human
involvement} and \textit{parameter update} and is thus much more cost-efficient
and time-saving than crowdsourcing. Experimental results on the MultiWOZ
dataset demonstrate that training a model on the simulated dialogues leads to
even better performance than using the same amount of human-generated dialogues
under the challenging low-resource settings, with as few as 85 dialogues as a
seed. When enough data is available, our method can still serve as an effective
data augmentation method. Human evaluation results also show that our simulated
dialogues have near-human fluency and annotation accuracy. The code and data
are available at \textbf{\url{https://github.com/Leezekun/dialogic}}.
| [
{
"version": "v1",
"created": "Sun, 9 Oct 2022 06:32:58 GMT"
},
{
"version": "v2",
"created": "Tue, 25 Oct 2022 03:10:53 GMT"
},
{
"version": "v3",
"created": "Sat, 12 Nov 2022 23:43:27 GMT"
},
{
"version": "v4",
"created": "Tue, 6 Jun 2023 02:19:08 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Li",
"Zekun",
""
],
[
"Chen",
"Wenhu",
""
],
[
"Li",
"Shiyang",
""
],
[
"Wang",
"Hong",
""
],
[
"Qian",
"Jing",
""
],
[
"Yan",
"Xifeng",
""
]
] | TITLE: Controllable Dialogue Simulation with In-Context Learning
ABSTRACT: Building dialogue systems requires a large corpus of annotated dialogues.
Such datasets are usually created via crowdsourcing, which is expensive and
time-consuming. In this paper, we propose \textsc{Dialogic}, a novel dialogue
simulation method based on large language model in-context learning to automate
dataset creation. Seeded with a few annotated dialogues, \textsc{Dialogic}
automatically selects in-context examples for demonstration and prompts GPT-3
to generate new dialogues and annotations in a controllable way. Our method can
rapidly expand a small set of dialogue data with minimum or zero \textit{human
involvement} and \textit{parameter update} and is thus much more cost-efficient
and time-saving than crowdsourcing. Experimental results on the MultiWOZ
dataset demonstrate that training a model on the simulated dialogues leads to
even better performance than using the same amount of human-generated dialogues
under the challenging low-resource settings, with as few as 85 dialogues as a
seed. When enough data is available, our method can still serve as an effective
data augmentation method. Human evaluation results also show that our simulated
dialogues have near-human fluency and annotation accuracy. The code and data
are available at \textbf{\url{https://github.com/Leezekun/dialogic}}.
|
2210.05241 | Chengting Yu | Chengting Yu, Zheming Gu, Da Li, Gaoang Wang, Aili Wang and Erping Li | STSC-SNN: Spatio-Temporal Synaptic Connection with Temporal Convolution
and Attention for Spiking Neural Networks | null | Frontiers in neuroscience, 2022, 12 | 10.3389/fnins.2022.1079357 | null | cs.NE q-bio.NC stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Spiking Neural Networks (SNNs), as one of the algorithmic models in
neuromorphic computing, have gained a great deal of research attention owing to
temporal information processing capability, low power consumption, and high
biological plausibility. The potential to efficiently extract spatio-temporal
features makes it suitable for processing the event streams. However, existing
synaptic structures in SNNs are almost full-connections or spatial 2D
convolution, neither of which can extract temporal dependencies adequately. In
this work, we take inspiration from biological synapses and propose a
spatio-temporal synaptic connection SNN (STSC-SNN) model, to enhance the
spatio-temporal receptive fields of synaptic connections, thereby establishing
temporal dependencies across layers. Concretely, we incorporate temporal
convolution and attention mechanisms to implement synaptic filtering and gating
functions. We show that endowing synaptic models with temporal dependencies can
improve the performance of SNNs on classification tasks. In addition, we
investigate the impact of performance vias varied spatial-temporal receptive
fields and reevaluate the temporal modules in SNNs. Our approach is tested on
neuromorphic datasets, including DVS128 Gesture (gesture recognition), N-MNIST,
CIFAR10-DVS (image classification), and SHD (speech digit recognition). The
results show that the proposed model outperforms the state-of-the-art accuracy
on nearly all datasets.
| [
{
"version": "v1",
"created": "Tue, 11 Oct 2022 08:13:22 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Yu",
"Chengting",
""
],
[
"Gu",
"Zheming",
""
],
[
"Li",
"Da",
""
],
[
"Wang",
"Gaoang",
""
],
[
"Wang",
"Aili",
""
],
[
"Li",
"Erping",
""
]
] | TITLE: STSC-SNN: Spatio-Temporal Synaptic Connection with Temporal Convolution
and Attention for Spiking Neural Networks
ABSTRACT: Spiking Neural Networks (SNNs), as one of the algorithmic models in
neuromorphic computing, have gained a great deal of research attention owing to
temporal information processing capability, low power consumption, and high
biological plausibility. The potential to efficiently extract spatio-temporal
features makes it suitable for processing the event streams. However, existing
synaptic structures in SNNs are almost full-connections or spatial 2D
convolution, neither of which can extract temporal dependencies adequately. In
this work, we take inspiration from biological synapses and propose a
spatio-temporal synaptic connection SNN (STSC-SNN) model, to enhance the
spatio-temporal receptive fields of synaptic connections, thereby establishing
temporal dependencies across layers. Concretely, we incorporate temporal
convolution and attention mechanisms to implement synaptic filtering and gating
functions. We show that endowing synaptic models with temporal dependencies can
improve the performance of SNNs on classification tasks. In addition, we
investigate the impact of performance vias varied spatial-temporal receptive
fields and reevaluate the temporal modules in SNNs. Our approach is tested on
neuromorphic datasets, including DVS128 Gesture (gesture recognition), N-MNIST,
CIFAR10-DVS (image classification), and SHD (speech digit recognition). The
results show that the proposed model outperforms the state-of-the-art accuracy
on nearly all datasets.
|
2210.07523 | Kosuke Nishida | Kosuke Nishida, Naoki Yoshinaga, Kyosuke Nishida | Self-Adaptive Named Entity Recognition by Retrieving Unstructured
Knowledge | EACL2023 (long) | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Although named entity recognition (NER) helps us to extract domain-specific
entities from text (e.g., artists in the music domain), it is costly to create
a large amount of training data or a structured knowledge base to perform
accurate NER in the target domain. Here, we propose self-adaptive NER, which
retrieves external knowledge from unstructured text to learn the usages of
entities that have not been learned well. To retrieve useful knowledge for NER,
we design an effective two-stage model that retrieves unstructured knowledge
using uncertain entities as queries. Our model predicts the entities in the
input and then finds those of which the prediction is not confident. Then, it
retrieves knowledge by using these uncertain entities as queries and
concatenates the retrieved text to the original input to revise the prediction.
Experiments on CrossNER datasets demonstrated that our model outperforms strong
baselines by 2.35 points in F1 metric.
| [
{
"version": "v1",
"created": "Fri, 14 Oct 2022 05:10:53 GMT"
},
{
"version": "v2",
"created": "Fri, 10 Mar 2023 12:32:16 GMT"
},
{
"version": "v3",
"created": "Tue, 6 Jun 2023 14:30:11 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Nishida",
"Kosuke",
""
],
[
"Yoshinaga",
"Naoki",
""
],
[
"Nishida",
"Kyosuke",
""
]
] | TITLE: Self-Adaptive Named Entity Recognition by Retrieving Unstructured
Knowledge
ABSTRACT: Although named entity recognition (NER) helps us to extract domain-specific
entities from text (e.g., artists in the music domain), it is costly to create
a large amount of training data or a structured knowledge base to perform
accurate NER in the target domain. Here, we propose self-adaptive NER, which
retrieves external knowledge from unstructured text to learn the usages of
entities that have not been learned well. To retrieve useful knowledge for NER,
we design an effective two-stage model that retrieves unstructured knowledge
using uncertain entities as queries. Our model predicts the entities in the
input and then finds those of which the prediction is not confident. Then, it
retrieves knowledge by using these uncertain entities as queries and
concatenates the retrieved text to the original input to revise the prediction.
Experiments on CrossNER datasets demonstrated that our model outperforms strong
baselines by 2.35 points in F1 metric.
|
2210.11466 | Yoonho Lee | Yoonho Lee, Annie S. Chen, Fahim Tajwar, Ananya Kumar, Huaxiu Yao,
Percy Liang, Chelsea Finn | Surgical Fine-Tuning Improves Adaptation to Distribution Shifts | ICLR 2023 | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A common approach to transfer learning under distribution shift is to
fine-tune the last few layers of a pre-trained model, preserving learned
features while also adapting to the new task. This paper shows that in such
settings, selectively fine-tuning a subset of layers (which we term surgical
fine-tuning) matches or outperforms commonly used fine-tuning approaches.
Moreover, the type of distribution shift influences which subset is more
effective to tune: for example, for image corruptions, fine-tuning only the
first few layers works best. We validate our findings systematically across
seven real-world data tasks spanning three types of distribution shifts.
Theoretically, we prove that for two-layer neural networks in an idealized
setting, first-layer tuning can outperform fine-tuning all layers. Intuitively,
fine-tuning more parameters on a small target dataset can cause information
learned during pre-training to be forgotten, and the relevant information
depends on the type of shift.
| [
{
"version": "v1",
"created": "Thu, 20 Oct 2022 17:59:15 GMT"
},
{
"version": "v2",
"created": "Thu, 2 Mar 2023 07:56:13 GMT"
},
{
"version": "v3",
"created": "Tue, 6 Jun 2023 05:58:11 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Lee",
"Yoonho",
""
],
[
"Chen",
"Annie S.",
""
],
[
"Tajwar",
"Fahim",
""
],
[
"Kumar",
"Ananya",
""
],
[
"Yao",
"Huaxiu",
""
],
[
"Liang",
"Percy",
""
],
[
"Finn",
"Chelsea",
""
]
] | TITLE: Surgical Fine-Tuning Improves Adaptation to Distribution Shifts
ABSTRACT: A common approach to transfer learning under distribution shift is to
fine-tune the last few layers of a pre-trained model, preserving learned
features while also adapting to the new task. This paper shows that in such
settings, selectively fine-tuning a subset of layers (which we term surgical
fine-tuning) matches or outperforms commonly used fine-tuning approaches.
Moreover, the type of distribution shift influences which subset is more
effective to tune: for example, for image corruptions, fine-tuning only the
first few layers works best. We validate our findings systematically across
seven real-world data tasks spanning three types of distribution shifts.
Theoretically, we prove that for two-layer neural networks in an idealized
setting, first-layer tuning can outperform fine-tuning all layers. Intuitively,
fine-tuning more parameters on a small target dataset can cause information
learned during pre-training to be forgotten, and the relevant information
depends on the type of shift.
|
2211.02139 | Faisal Hamman | Faisal Hamman, Jiahao Chen, Sanghamitra Dutta | Can Querying for Bias Leak Protected Attributes? Achieving Privacy With
Smooth Sensitivity | Published in 2023 ACM Conference on Fairness, Accountability, and
Transparency (FAccT2023) | null | 10.1145/3593013.3594086 | null | cs.LG cs.AI cs.CR cs.CY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Existing regulations prohibit model developers from accessing protected
attributes (gender, race, etc.), often resulting in fairness assessments on
populations without knowing their protected groups. In such scenarios,
institutions often adopt a separation between the model developers (who train
models with no access to the protected attributes) and a compliance team (who
may have access to the entire dataset for auditing purposes). However, the
model developers might be allowed to test their models for bias by querying the
compliance team for group fairness metrics. In this paper, we first demonstrate
that simply querying for fairness metrics, such as statistical parity and
equalized odds can leak the protected attributes of individuals to the model
developers. We demonstrate that there always exist strategies by which the
model developers can identify the protected attribute of a targeted individual
in the test dataset from just a single query. In particular, we show that one
can reconstruct the protected attributes of all the individuals from O(Nk \log(
n /Nk)) queries when Nk<<n using techniques from compressed sensing (n: size of
the test dataset, Nk: size of smallest group). Our results pose an interesting
debate in algorithmic fairness: should querying for fairness metrics be viewed
as a neutral-valued solution to ensure compliance with regulations? Or, does it
constitute a violation of regulations and privacy if the number of queries
answered is enough for the model developers to identify the protected
attributes of specific individuals? To address this supposed violation, we also
propose Attribute-Conceal, a novel technique that achieves differential privacy
by calibrating noise to the smooth sensitivity of our bias query, outperforming
naive techniques such as the Laplace mechanism. We also include experimental
results on the Adult dataset and synthetic data.
| [
{
"version": "v1",
"created": "Thu, 3 Nov 2022 20:44:48 GMT"
},
{
"version": "v2",
"created": "Mon, 5 Jun 2023 20:55:12 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Hamman",
"Faisal",
""
],
[
"Chen",
"Jiahao",
""
],
[
"Dutta",
"Sanghamitra",
""
]
] | TITLE: Can Querying for Bias Leak Protected Attributes? Achieving Privacy With
Smooth Sensitivity
ABSTRACT: Existing regulations prohibit model developers from accessing protected
attributes (gender, race, etc.), often resulting in fairness assessments on
populations without knowing their protected groups. In such scenarios,
institutions often adopt a separation between the model developers (who train
models with no access to the protected attributes) and a compliance team (who
may have access to the entire dataset for auditing purposes). However, the
model developers might be allowed to test their models for bias by querying the
compliance team for group fairness metrics. In this paper, we first demonstrate
that simply querying for fairness metrics, such as statistical parity and
equalized odds can leak the protected attributes of individuals to the model
developers. We demonstrate that there always exist strategies by which the
model developers can identify the protected attribute of a targeted individual
in the test dataset from just a single query. In particular, we show that one
can reconstruct the protected attributes of all the individuals from O(Nk \log(
n /Nk)) queries when Nk<<n using techniques from compressed sensing (n: size of
the test dataset, Nk: size of smallest group). Our results pose an interesting
debate in algorithmic fairness: should querying for fairness metrics be viewed
as a neutral-valued solution to ensure compliance with regulations? Or, does it
constitute a violation of regulations and privacy if the number of queries
answered is enough for the model developers to identify the protected
attributes of specific individuals? To address this supposed violation, we also
propose Attribute-Conceal, a novel technique that achieves differential privacy
by calibrating noise to the smooth sensitivity of our bias query, outperforming
naive techniques such as the Laplace mechanism. We also include experimental
results on the Adult dataset and synthetic data.
|
2211.04022 | Yinghui He | Yinghui He, Guanding Yu, Yunlong Cai, and Haiyan Luo | Integrated Sensing, Computation, and Communication: System Framework and
Performance Optimization | null | null | null | null | cs.IT math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Integrated sensing, computation, and communication (ISCC) has been recently
considered as a promising technique for beyond 5G systems. In ISCC systems, the
competition for communication and computation resources between sensing tasks
for ambient intelligence and computation tasks from mobile devices becomes an
increasingly challenging issue. To address it, we first propose an efficient
sensing framework with a novel action detection module. In this module, a
threshold is used for detecting whether the sensing target is static and thus
the overhead can be reduced. Subsequently, we mathematically analyze the
sensing performance of the proposed framework and theoretically prove its
effectiveness with the help of the sampling theorem. Based on sensing
performance models, we formulate a sensing performance maximization problem
while guaranteeing the quality-of-service (QoS) requirements of tasks. To solve
it, we propose an optimal resource allocation strategy, in which the minimum
resource is allocated to computation tasks, and the rest is devoted to the
sensing task. Besides, a threshold selection policy is derived and the results
further demonstrate the necessity of the proposed sensing framework. Finally, a
real-world test of action recognition tasks based on USRP B210 is conducted to
verify the sensing performance analysis. Extensive experiments demonstrate the
performance improvement of our proposal by comparing it with some benchmark
schemes.
| [
{
"version": "v1",
"created": "Tue, 8 Nov 2022 05:55:52 GMT"
},
{
"version": "v2",
"created": "Wed, 9 Nov 2022 10:58:48 GMT"
},
{
"version": "v3",
"created": "Tue, 6 Jun 2023 11:46:14 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"He",
"Yinghui",
""
],
[
"Yu",
"Guanding",
""
],
[
"Cai",
"Yunlong",
""
],
[
"Luo",
"Haiyan",
""
]
] | TITLE: Integrated Sensing, Computation, and Communication: System Framework and
Performance Optimization
ABSTRACT: Integrated sensing, computation, and communication (ISCC) has been recently
considered as a promising technique for beyond 5G systems. In ISCC systems, the
competition for communication and computation resources between sensing tasks
for ambient intelligence and computation tasks from mobile devices becomes an
increasingly challenging issue. To address it, we first propose an efficient
sensing framework with a novel action detection module. In this module, a
threshold is used for detecting whether the sensing target is static and thus
the overhead can be reduced. Subsequently, we mathematically analyze the
sensing performance of the proposed framework and theoretically prove its
effectiveness with the help of the sampling theorem. Based on sensing
performance models, we formulate a sensing performance maximization problem
while guaranteeing the quality-of-service (QoS) requirements of tasks. To solve
it, we propose an optimal resource allocation strategy, in which the minimum
resource is allocated to computation tasks, and the rest is devoted to the
sensing task. Besides, a threshold selection policy is derived and the results
further demonstrate the necessity of the proposed sensing framework. Finally, a
real-world test of action recognition tasks based on USRP B210 is conducted to
verify the sensing performance analysis. Extensive experiments demonstrate the
performance improvement of our proposal by comparing it with some benchmark
schemes.
|
2211.08099 | Haike Xu | Haike Xu, Zongyu Lin, Jing Zhou, Yanan Zheng, Zhilin Yang | A Universal Discriminator for Zero-Shot Generalization | ACL 2023 main conference (Long paper) | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Generative modeling has been the dominant approach for large-scale
pretraining and zero-shot generalization. In this work, we challenge this
convention by showing that discriminative approaches perform substantially
better than generative ones on a large number of NLP tasks. Technically, we
train a single discriminator to predict whether a text sample comes from the
true data distribution, similar to GANs. Since many NLP tasks can be formulated
as selecting from a few options, we use this discriminator to predict the
concatenation of input and which option has the highest probability of coming
from the true data distribution. This simple formulation achieves
state-of-the-art zero-shot results on the T0 benchmark, outperforming T0 by
16.0\%, 7.8\%, and 11.5\% respectively on different scales. In the finetuning
setting, our approach also achieves new state-of-the-art results on a wide
range of NLP tasks, with only 1/4 parameters of previous methods. Meanwhile,
our approach requires minimal prompting efforts, which largely improves
robustness and is essential for real-world applications. Furthermore, we also
jointly train a generalized UD in combination with generative tasks, which
maintains its advantage on discriminative tasks and simultaneously works on
generative tasks.
| [
{
"version": "v1",
"created": "Tue, 15 Nov 2022 12:33:31 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 03:01:43 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Xu",
"Haike",
""
],
[
"Lin",
"Zongyu",
""
],
[
"Zhou",
"Jing",
""
],
[
"Zheng",
"Yanan",
""
],
[
"Yang",
"Zhilin",
""
]
] | TITLE: A Universal Discriminator for Zero-Shot Generalization
ABSTRACT: Generative modeling has been the dominant approach for large-scale
pretraining and zero-shot generalization. In this work, we challenge this
convention by showing that discriminative approaches perform substantially
better than generative ones on a large number of NLP tasks. Technically, we
train a single discriminator to predict whether a text sample comes from the
true data distribution, similar to GANs. Since many NLP tasks can be formulated
as selecting from a few options, we use this discriminator to predict the
concatenation of input and which option has the highest probability of coming
from the true data distribution. This simple formulation achieves
state-of-the-art zero-shot results on the T0 benchmark, outperforming T0 by
16.0\%, 7.8\%, and 11.5\% respectively on different scales. In the finetuning
setting, our approach also achieves new state-of-the-art results on a wide
range of NLP tasks, with only 1/4 parameters of previous methods. Meanwhile,
our approach requires minimal prompting efforts, which largely improves
robustness and is essential for real-world applications. Furthermore, we also
jointly train a generalized UD in combination with generative tasks, which
maintains its advantage on discriminative tasks and simultaneously works on
generative tasks.
|
2211.08583 | Hiroki Naganuma | Hiroki Naganuma, Kartik Ahuja, Shiro Takagi, Tetsuya Motokawa, Rio
Yokota, Kohta Ishikawa, Ikuro Sato, Ioannis Mitliagkas | Empirical Study on Optimizer Selection for Out-of-Distribution
Generalization | Accepted to TMLR | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Modern deep learning systems do not generalize well when the test data
distribution is slightly different to the training data distribution. While
much promising work has been accomplished to address this fragility, a
systematic study of the role of optimizers and their out-of-distribution
generalization performance has not been undertaken. In this study, we examine
the performance of popular first-order optimizers for different classes of
distributional shift under empirical risk minimization and invariant risk
minimization. We address this question for image and text classification using
DomainBed, WILDS, and Backgrounds Challenge as testbeds for studying different
types of shifts -- namely correlation and diversity shift. We search over a
wide range of hyperparameters and examine classification accuracy
(in-distribution and out-of-distribution) for over 20,000 models. We arrive at
the following findings, which we expect to be helpful for practitioners: i)
adaptive optimizers (e.g., Adam) perform worse than non-adaptive optimizers
(e.g., SGD, momentum SGD) on out-of-distribution performance. In particular,
even though there is no significant difference in in-distribution performance,
we show a measurable difference in out-of-distribution performance. ii)
in-distribution performance and out-of-distribution performance exhibit three
types of behavior depending on the dataset -- linear returns, increasing
returns, and diminishing returns. For example, in the training of natural
language data using Adam, fine-tuning the performance of in-distribution
performance does not significantly contribute to the out-of-distribution
generalization performance.
| [
{
"version": "v1",
"created": "Tue, 15 Nov 2022 23:56:30 GMT"
},
{
"version": "v2",
"created": "Fri, 18 Nov 2022 21:29:02 GMT"
},
{
"version": "v3",
"created": "Mon, 5 Jun 2023 22:23:52 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Naganuma",
"Hiroki",
""
],
[
"Ahuja",
"Kartik",
""
],
[
"Takagi",
"Shiro",
""
],
[
"Motokawa",
"Tetsuya",
""
],
[
"Yokota",
"Rio",
""
],
[
"Ishikawa",
"Kohta",
""
],
[
"Sato",
"Ikuro",
""
],
[
"Mitliagkas",
"Ioannis",
""
]
] | TITLE: Empirical Study on Optimizer Selection for Out-of-Distribution
Generalization
ABSTRACT: Modern deep learning systems do not generalize well when the test data
distribution is slightly different to the training data distribution. While
much promising work has been accomplished to address this fragility, a
systematic study of the role of optimizers and their out-of-distribution
generalization performance has not been undertaken. In this study, we examine
the performance of popular first-order optimizers for different classes of
distributional shift under empirical risk minimization and invariant risk
minimization. We address this question for image and text classification using
DomainBed, WILDS, and Backgrounds Challenge as testbeds for studying different
types of shifts -- namely correlation and diversity shift. We search over a
wide range of hyperparameters and examine classification accuracy
(in-distribution and out-of-distribution) for over 20,000 models. We arrive at
the following findings, which we expect to be helpful for practitioners: i)
adaptive optimizers (e.g., Adam) perform worse than non-adaptive optimizers
(e.g., SGD, momentum SGD) on out-of-distribution performance. In particular,
even though there is no significant difference in in-distribution performance,
we show a measurable difference in out-of-distribution performance. ii)
in-distribution performance and out-of-distribution performance exhibit three
types of behavior depending on the dataset -- linear returns, increasing
returns, and diminishing returns. For example, in the training of natural
language data using Adam, fine-tuning the performance of in-distribution
performance does not significantly contribute to the out-of-distribution
generalization performance.
|
2211.12561 | Michihiro Yasunaga | Michihiro Yasunaga, Armen Aghajanyan, Weijia Shi, Rich James, Jure
Leskovec, Percy Liang, Mike Lewis, Luke Zettlemoyer, Wen-tau Yih | Retrieval-Augmented Multimodal Language Modeling | Published at ICML 2023. Blog post available at
https://cs.stanford.edu/~myasu/blog/racm3/ | null | null | null | cs.CV cs.CL cs.LG | http://creativecommons.org/licenses/by/4.0/ | Recent multimodal models such as DALL-E and CM3 have achieved remarkable
progress in text-to-image and image-to-text generation. However, these models
store all learned knowledge (e.g., the appearance of the Eiffel Tower) in the
model parameters, requiring increasingly larger models and training data to
capture more knowledge. To integrate knowledge in a more scalable and modular
way, we propose a retrieval-augmented multimodal model, which enables a base
multimodal model (generator) to refer to relevant text and images fetched by a
retriever from external memory (e.g., documents on the web). Specifically, for
the retriever, we use a pretrained CLIP, and for the generator, we train a CM3
Transformer on the LAION dataset. Our resulting model, named
Retrieval-Augmented CM3 (RA-CM3), is the first multimodal model that can
retrieve and generate both text and images. We show that RA-CM3 significantly
outperforms baseline multimodal models such as DALL-E and CM3 on both image and
caption generation tasks (12 FID and 17 CIDEr improvements on MS-COCO), while
requiring much less compute for training (<30% of DALL-E). Moreover, we show
that RA-CM3 exhibits novel capabilities, such as faithful image generation and
multimodal in-context learning (e.g., image generation from demonstrations).
| [
{
"version": "v1",
"created": "Tue, 22 Nov 2022 20:26:44 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 00:28:34 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Yasunaga",
"Michihiro",
""
],
[
"Aghajanyan",
"Armen",
""
],
[
"Shi",
"Weijia",
""
],
[
"James",
"Rich",
""
],
[
"Leskovec",
"Jure",
""
],
[
"Liang",
"Percy",
""
],
[
"Lewis",
"Mike",
""
],
[
"Zettlemoyer",
"Luke",
""
],
[
"Yih",
"Wen-tau",
""
]
] | TITLE: Retrieval-Augmented Multimodal Language Modeling
ABSTRACT: Recent multimodal models such as DALL-E and CM3 have achieved remarkable
progress in text-to-image and image-to-text generation. However, these models
store all learned knowledge (e.g., the appearance of the Eiffel Tower) in the
model parameters, requiring increasingly larger models and training data to
capture more knowledge. To integrate knowledge in a more scalable and modular
way, we propose a retrieval-augmented multimodal model, which enables a base
multimodal model (generator) to refer to relevant text and images fetched by a
retriever from external memory (e.g., documents on the web). Specifically, for
the retriever, we use a pretrained CLIP, and for the generator, we train a CM3
Transformer on the LAION dataset. Our resulting model, named
Retrieval-Augmented CM3 (RA-CM3), is the first multimodal model that can
retrieve and generate both text and images. We show that RA-CM3 significantly
outperforms baseline multimodal models such as DALL-E and CM3 on both image and
caption generation tasks (12 FID and 17 CIDEr improvements on MS-COCO), while
requiring much less compute for training (<30% of DALL-E). Moreover, we show
that RA-CM3 exhibits novel capabilities, such as faithful image generation and
multimodal in-context learning (e.g., image generation from demonstrations).
|
2212.03178 | Mohsen Hooshmand | Alireza Abdi, Masih Hajsaeedi, Mohsen Hooshmand | Longest Common Substring in Longest Common Subsequence's Solution
Service: A Novel Hyper-Heuristic | null | null | 10.1016/j.compbiolchem.2023.107882 | null | cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Longest Common Subsequence (LCS) is the problem of finding a subsequence
among a set of strings that has two properties of being common to all and is
the longest. The LCS has applications in computational biology and text
editing, among many others. Due to the NP-hardness of the general longest
common subsequence, numerous heuristic algorithms and solvers have been
proposed to give the best possible solution for different sets of strings. None
of them has the best performance for all types of sets. In addition, there is
no method to specify the type of a given set of strings. Besides that, the
available hyper-heuristic is not efficient and fast enough to solve this
problem in real-world applications. This paper proposes a novel hyper-heuristic
to solve the longest common subsequence problem using a novel criterion to
classify a set of strings based on their similarity. To do this, we offer a
general stochastic framework to identify the type of a given set of strings.
Following that, we introduce the set similarity dichotomizer ($S^2D$) algorithm
based on the framework that divides the type of sets into two. This algorithm
is introduced for the first time in this paper and opens a new way to go beyond
the current LCS solvers. Then, we present a novel hyper-heuristic that exploits
the $S^2D$ and one of the internal properties of the set to choose the best
matching heuristic among a set of heuristics. We compare the results on
benchmark datasets with the best heuristics and hyper-heuristics. The results
show a higher performance of our proposed hyper-heuristic in both quality of
solutions and run time factors.
| [
{
"version": "v1",
"created": "Sat, 3 Dec 2022 07:52:57 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Abdi",
"Alireza",
""
],
[
"Hajsaeedi",
"Masih",
""
],
[
"Hooshmand",
"Mohsen",
""
]
] | TITLE: Longest Common Substring in Longest Common Subsequence's Solution
Service: A Novel Hyper-Heuristic
ABSTRACT: The Longest Common Subsequence (LCS) is the problem of finding a subsequence
among a set of strings that has two properties of being common to all and is
the longest. The LCS has applications in computational biology and text
editing, among many others. Due to the NP-hardness of the general longest
common subsequence, numerous heuristic algorithms and solvers have been
proposed to give the best possible solution for different sets of strings. None
of them has the best performance for all types of sets. In addition, there is
no method to specify the type of a given set of strings. Besides that, the
available hyper-heuristic is not efficient and fast enough to solve this
problem in real-world applications. This paper proposes a novel hyper-heuristic
to solve the longest common subsequence problem using a novel criterion to
classify a set of strings based on their similarity. To do this, we offer a
general stochastic framework to identify the type of a given set of strings.
Following that, we introduce the set similarity dichotomizer ($S^2D$) algorithm
based on the framework that divides the type of sets into two. This algorithm
is introduced for the first time in this paper and opens a new way to go beyond
the current LCS solvers. Then, we present a novel hyper-heuristic that exploits
the $S^2D$ and one of the internal properties of the set to choose the best
matching heuristic among a set of heuristics. We compare the results on
benchmark datasets with the best heuristics and hyper-heuristics. The results
show a higher performance of our proposed hyper-heuristic in both quality of
solutions and run time factors.
|
2212.07981 | Yixin Liu | Yixin Liu, Alexander R. Fabbri, Pengfei Liu, Yilun Zhao, Linyong Nan,
Ruilin Han, Simeng Han, Shafiq Joty, Chien-Sheng Wu, Caiming Xiong, Dragomir
Radev | Revisiting the Gold Standard: Grounding Summarization Evaluation with
Robust Human Evaluation | ACL 2023 Camera Ready | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Human evaluation is the foundation upon which the evaluation of both
summarization systems and automatic metrics rests. However, existing human
evaluation studies for summarization either exhibit a low inter-annotator
agreement or have insufficient scale, and an in-depth analysis of human
evaluation is lacking. Therefore, we address the shortcomings of existing
summarization evaluation along the following axes: (1) We propose a modified
summarization salience protocol, Atomic Content Units (ACUs), which is based on
fine-grained semantic units and allows for a high inter-annotator agreement.
(2) We curate the Robust Summarization Evaluation (RoSE) benchmark, a large
human evaluation dataset consisting of 22,000 summary-level annotations over 28
top-performing systems on three datasets. (3) We conduct a comparative study of
four human evaluation protocols, underscoring potential confounding factors in
evaluation setups. (4) We evaluate 50 automatic metrics and their variants
using the collected human annotations across evaluation protocols and
demonstrate how our benchmark leads to more statistically stable and
significant results. The metrics we benchmarked include recent methods based on
large language models (LLMs), GPTScore and G-Eval. Furthermore, our findings
have important implications for evaluating LLMs, as we show that LLMs adjusted
by human feedback (e.g., GPT-3.5) may overfit unconstrained human evaluation,
which is affected by the annotators' prior, input-agnostic preferences, calling
for more robust, targeted evaluation methods.
| [
{
"version": "v1",
"created": "Thu, 15 Dec 2022 17:26:05 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 07:45:27 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Liu",
"Yixin",
""
],
[
"Fabbri",
"Alexander R.",
""
],
[
"Liu",
"Pengfei",
""
],
[
"Zhao",
"Yilun",
""
],
[
"Nan",
"Linyong",
""
],
[
"Han",
"Ruilin",
""
],
[
"Han",
"Simeng",
""
],
[
"Joty",
"Shafiq",
""
],
[
"Wu",
"Chien-Sheng",
""
],
[
"Xiong",
"Caiming",
""
],
[
"Radev",
"Dragomir",
""
]
] | TITLE: Revisiting the Gold Standard: Grounding Summarization Evaluation with
Robust Human Evaluation
ABSTRACT: Human evaluation is the foundation upon which the evaluation of both
summarization systems and automatic metrics rests. However, existing human
evaluation studies for summarization either exhibit a low inter-annotator
agreement or have insufficient scale, and an in-depth analysis of human
evaluation is lacking. Therefore, we address the shortcomings of existing
summarization evaluation along the following axes: (1) We propose a modified
summarization salience protocol, Atomic Content Units (ACUs), which is based on
fine-grained semantic units and allows for a high inter-annotator agreement.
(2) We curate the Robust Summarization Evaluation (RoSE) benchmark, a large
human evaluation dataset consisting of 22,000 summary-level annotations over 28
top-performing systems on three datasets. (3) We conduct a comparative study of
four human evaluation protocols, underscoring potential confounding factors in
evaluation setups. (4) We evaluate 50 automatic metrics and their variants
using the collected human annotations across evaluation protocols and
demonstrate how our benchmark leads to more statistically stable and
significant results. The metrics we benchmarked include recent methods based on
large language models (LLMs), GPTScore and G-Eval. Furthermore, our findings
have important implications for evaluating LLMs, as we show that LLMs adjusted
by human feedback (e.g., GPT-3.5) may overfit unconstrained human evaluation,
which is affected by the annotators' prior, input-agnostic preferences, calling
for more robust, targeted evaluation methods.
|
2212.12139 | Fucai Ke | Fucai Ke, Weiqing Wang, Weicong Tan, Lan Du, Yuan Jin, Yujin Huang and
Hongzhi Yin | HiTSKT: A Hierarchical Transformer Model for Session-Aware Knowledge
Tracing | null | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | Knowledge tracing (KT) aims to leverage students' learning histories to
estimate their mastery levels on a set of pre-defined skills, based on which
the corresponding future performance can be accurately predicted. As an
important way of providing personalized experience for online education, KT has
gained increased attention in recent years. In practice, a student's learning
history comprises answers to sets of massed questions, each known as a session,
rather than merely being a sequence of independent answers. Theoretically,
within and across these sessions, students' learning dynamics can be very
different. Therefore, how to effectively model the dynamics of students'
knowledge states within and across the sessions is crucial for handling the KT
problem. Most existing KT models treat student's learning records as a single
continuing sequence, without capturing the sessional shift of students'
knowledge state. To address the above issue, we propose a novel hierarchical
transformer model, named HiTSKT, comprises an interaction(-level) encoder to
capture the knowledge a student acquires within a session, and a
session(-level) encoder to summarise acquired knowledge across the past
sessions. To predict an interaction in the current session, a knowledge
retriever integrates the summarised past-session knowledge with the previous
interactions' information into proper knowledge representations. These
representations are then used to compute the student's current knowledge state.
Additionally, to model the student's long-term forgetting behaviour across the
sessions, a power-law-decay attention mechanism is designed and deployed in the
session encoder, allowing it to emphasize more on the recent sessions.
Extensive experiments on three public datasets demonstrate that HiTSKT achieves
new state-of-the-art performance on all the datasets compared with six
state-of-the-art KT models.
| [
{
"version": "v1",
"created": "Fri, 23 Dec 2022 04:22:42 GMT"
},
{
"version": "v2",
"created": "Thu, 12 Jan 2023 12:52:16 GMT"
},
{
"version": "v3",
"created": "Tue, 6 Jun 2023 13:05:01 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Ke",
"Fucai",
""
],
[
"Wang",
"Weiqing",
""
],
[
"Tan",
"Weicong",
""
],
[
"Du",
"Lan",
""
],
[
"Jin",
"Yuan",
""
],
[
"Huang",
"Yujin",
""
],
[
"Yin",
"Hongzhi",
""
]
] | TITLE: HiTSKT: A Hierarchical Transformer Model for Session-Aware Knowledge
Tracing
ABSTRACT: Knowledge tracing (KT) aims to leverage students' learning histories to
estimate their mastery levels on a set of pre-defined skills, based on which
the corresponding future performance can be accurately predicted. As an
important way of providing personalized experience for online education, KT has
gained increased attention in recent years. In practice, a student's learning
history comprises answers to sets of massed questions, each known as a session,
rather than merely being a sequence of independent answers. Theoretically,
within and across these sessions, students' learning dynamics can be very
different. Therefore, how to effectively model the dynamics of students'
knowledge states within and across the sessions is crucial for handling the KT
problem. Most existing KT models treat student's learning records as a single
continuing sequence, without capturing the sessional shift of students'
knowledge state. To address the above issue, we propose a novel hierarchical
transformer model, named HiTSKT, comprises an interaction(-level) encoder to
capture the knowledge a student acquires within a session, and a
session(-level) encoder to summarise acquired knowledge across the past
sessions. To predict an interaction in the current session, a knowledge
retriever integrates the summarised past-session knowledge with the previous
interactions' information into proper knowledge representations. These
representations are then used to compute the student's current knowledge state.
Additionally, to model the student's long-term forgetting behaviour across the
sessions, a power-law-decay attention mechanism is designed and deployed in the
session encoder, allowing it to emphasize more on the recent sessions.
Extensive experiments on three public datasets demonstrate that HiTSKT achieves
new state-of-the-art performance on all the datasets compared with six
state-of-the-art KT models.
|
2301.00489 | Jiayun Zhang | Jiayun Zhang, Xiyuan Zhang, Xinyang Zhang, Dezhi Hong, Rajesh K.
Gupta, Jingbo Shang | Navigating Alignment for Non-identical Client Class Sets: A Label
Name-Anchored Federated Learning Framework | Accepted by KDD 2023 | null | 10.1145/3580305.3599443 | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Traditional federated classification methods, even those designed for non-IID
clients, assume that each client annotates its local data with respect to the
same universal class set. In this paper, we focus on a more general yet
practical setting, non-identical client class sets, where clients focus on
their own (different or even non-overlapping) class sets and seek a global
model that works for the union of these classes. If one views classification as
finding the best match between representations produced by data/label encoder,
such heterogeneity in client class sets poses a new significant challenge --
local encoders at different clients may operate in different and even
independent latent spaces, making it hard to aggregate at the server. We
propose a novel framework, FedAlign, to align the latent spaces across clients
from both label and data perspectives. From a label perspective, we leverage
the expressive natural language class names as a common ground for label
encoders to anchor class representations and guide the data encoder learning
across clients. From a data perspective, during local training, we regard the
global class representations as anchors and leverage the data points that are
close/far enough to the anchors of locally-unaware classes to align the data
encoders across clients. Our theoretical analysis of the generalization
performance and extensive experiments on four real-world datasets of different
tasks confirm that FedAlign outperforms various state-of-the-art (non-IID)
federated classification methods.
| [
{
"version": "v1",
"created": "Sun, 1 Jan 2023 23:17:30 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 05:30:05 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Zhang",
"Jiayun",
""
],
[
"Zhang",
"Xiyuan",
""
],
[
"Zhang",
"Xinyang",
""
],
[
"Hong",
"Dezhi",
""
],
[
"Gupta",
"Rajesh K.",
""
],
[
"Shang",
"Jingbo",
""
]
] | TITLE: Navigating Alignment for Non-identical Client Class Sets: A Label
Name-Anchored Federated Learning Framework
ABSTRACT: Traditional federated classification methods, even those designed for non-IID
clients, assume that each client annotates its local data with respect to the
same universal class set. In this paper, we focus on a more general yet
practical setting, non-identical client class sets, where clients focus on
their own (different or even non-overlapping) class sets and seek a global
model that works for the union of these classes. If one views classification as
finding the best match between representations produced by data/label encoder,
such heterogeneity in client class sets poses a new significant challenge --
local encoders at different clients may operate in different and even
independent latent spaces, making it hard to aggregate at the server. We
propose a novel framework, FedAlign, to align the latent spaces across clients
from both label and data perspectives. From a label perspective, we leverage
the expressive natural language class names as a common ground for label
encoders to anchor class representations and guide the data encoder learning
across clients. From a data perspective, during local training, we regard the
global class representations as anchors and leverage the data points that are
close/far enough to the anchors of locally-unaware classes to align the data
encoders across clients. Our theoretical analysis of the generalization
performance and extensive experiments on four real-world datasets of different
tasks confirm that FedAlign outperforms various state-of-the-art (non-IID)
federated classification methods.
|
2301.03722 | Argha Sen | Argha Sen, Ayan Zunaid, Soumyajit Chatterjee, Basabdatta Palit, Sandip
Chakraborty | Revisiting Cellular Throughput Prediction over the Edge: Collaborative
Multi-device, Multi-network in-situ Learning | 12 pages, 10 figures, Accepted in International Conference on
Embedded Wireless Systems and Networks (EWSN 2023) | null | null | null | cs.NI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Pervasive applications over large-scale, distributed embedded devices and the
Internet of Things (IoT) demand precise coordination with the network; for
example, several such applications, like collaborative video streaming and live
analysis, augmented reality, etc., need continuous monitoring of network
throughput and adapt the application behavior accordingly. Although the idea of
network throughput prediction is not new and quite dated, in this paper, we
show that the existing approaches fail to correctly infer the throughput when
the network operator or the device change, and thus, not generic enough for
Internet-scale applications. We propose \ourmethod, a novel approach that
allows collaborative training across different client hardware by capturing
throughput variations based on devices' sensitivity towards the corresponding
network configurations. Rigorous evaluations show that \ourmethod{} outperforms
various standard baseline algorithms with more than $80\%$ R2-score over
different datasets. We also analyze the performance of \ourmethod{} over a
network-aware streaming media application and demonstrate its efficacy for
various application scenarios.
| [
{
"version": "v1",
"created": "Mon, 9 Jan 2023 23:50:41 GMT"
},
{
"version": "v2",
"created": "Mon, 5 Jun 2023 19:10:27 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Sen",
"Argha",
""
],
[
"Zunaid",
"Ayan",
""
],
[
"Chatterjee",
"Soumyajit",
""
],
[
"Palit",
"Basabdatta",
""
],
[
"Chakraborty",
"Sandip",
""
]
] | TITLE: Revisiting Cellular Throughput Prediction over the Edge: Collaborative
Multi-device, Multi-network in-situ Learning
ABSTRACT: Pervasive applications over large-scale, distributed embedded devices and the
Internet of Things (IoT) demand precise coordination with the network; for
example, several such applications, like collaborative video streaming and live
analysis, augmented reality, etc., need continuous monitoring of network
throughput and adapt the application behavior accordingly. Although the idea of
network throughput prediction is not new and quite dated, in this paper, we
show that the existing approaches fail to correctly infer the throughput when
the network operator or the device change, and thus, not generic enough for
Internet-scale applications. We propose \ourmethod, a novel approach that
allows collaborative training across different client hardware by capturing
throughput variations based on devices' sensitivity towards the corresponding
network configurations. Rigorous evaluations show that \ourmethod{} outperforms
various standard baseline algorithms with more than $80\%$ R2-score over
different datasets. We also analyze the performance of \ourmethod{} over a
network-aware streaming media application and demonstrate its efficacy for
various application scenarios.
|
2301.13293 | Rishabh Tiwari | Rishabh Tiwari, Pradeep Shenoy | Overcoming Simplicity Bias in Deep Networks using a Feature Sieve | Accepted at ICML 2023 | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Simplicity bias is the concerning tendency of deep networks to over-depend on
simple, weakly predictive features, to the exclusion of stronger, more complex
features. This is exacerbated in real-world applications by limited training
data and spurious feature-label correlations, leading to biased, incorrect
predictions. We propose a direct, interventional method for addressing
simplicity bias in DNNs, which we call the feature sieve. We aim to
automatically identify and suppress easily-computable spurious features in
lower layers of the network, thereby allowing the higher network levels to
extract and utilize richer, more meaningful representations. We provide
concrete evidence of this differential suppression & enhancement of relevant
features on both controlled datasets and real-world images, and report
substantial gains on many real-world debiasing benchmarks (11.4% relative gain
on Imagenet-A; 3.2% on BAR, etc). Crucially, we do not depend on prior
knowledge of spurious attributes or features, and in fact outperform many
baselines that explicitly incorporate such information. We believe that our
feature sieve work opens up exciting new research directions in automated
adversarial feature extraction and representation learning for deep networks.
| [
{
"version": "v1",
"created": "Mon, 30 Jan 2023 21:11:13 GMT"
},
{
"version": "v2",
"created": "Thu, 2 Feb 2023 11:06:26 GMT"
},
{
"version": "v3",
"created": "Tue, 6 Jun 2023 16:45:31 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Tiwari",
"Rishabh",
""
],
[
"Shenoy",
"Pradeep",
""
]
] | TITLE: Overcoming Simplicity Bias in Deep Networks using a Feature Sieve
ABSTRACT: Simplicity bias is the concerning tendency of deep networks to over-depend on
simple, weakly predictive features, to the exclusion of stronger, more complex
features. This is exacerbated in real-world applications by limited training
data and spurious feature-label correlations, leading to biased, incorrect
predictions. We propose a direct, interventional method for addressing
simplicity bias in DNNs, which we call the feature sieve. We aim to
automatically identify and suppress easily-computable spurious features in
lower layers of the network, thereby allowing the higher network levels to
extract and utilize richer, more meaningful representations. We provide
concrete evidence of this differential suppression & enhancement of relevant
features on both controlled datasets and real-world images, and report
substantial gains on many real-world debiasing benchmarks (11.4% relative gain
on Imagenet-A; 3.2% on BAR, etc). Crucially, we do not depend on prior
knowledge of spurious attributes or features, and in fact outperform many
baselines that explicitly incorporate such information. We believe that our
feature sieve work opens up exciting new research directions in automated
adversarial feature extraction and representation learning for deep networks.
|
2302.00093 | Xinyun Chen | Freda Shi, Xinyun Chen, Kanishka Misra, Nathan Scales, David Dohan, Ed
Chi, Nathanael Sch\"arli, Denny Zhou | Large Language Models Can Be Easily Distracted by Irrelevant Context | Published in ICML 2023 | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Large language models have achieved impressive performance on various natural
language processing tasks. However, so far they have been evaluated primarily
on benchmarks where all information in the input context is relevant for
solving the task. In this work, we investigate the distractibility of large
language models, i.e., how the model problem-solving accuracy can be influenced
by irrelevant context. In particular, we introduce Grade-School Math with
Irrelevant Context (GSM-IC), an arithmetic reasoning dataset with irrelevant
information in the problem description. We use this benchmark to measure the
distractibility of cutting-edge prompting techniques for large language models,
and find that the model performance is dramatically decreased when irrelevant
information is included. We also identify several approaches for mitigating
this deficiency, such as decoding with self-consistency and adding to the
prompt an instruction that tells the language model to ignore the irrelevant
information.
| [
{
"version": "v1",
"created": "Tue, 31 Jan 2023 20:48:57 GMT"
},
{
"version": "v2",
"created": "Mon, 13 Feb 2023 20:08:59 GMT"
},
{
"version": "v3",
"created": "Tue, 6 Jun 2023 08:36:20 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Shi",
"Freda",
""
],
[
"Chen",
"Xinyun",
""
],
[
"Misra",
"Kanishka",
""
],
[
"Scales",
"Nathan",
""
],
[
"Dohan",
"David",
""
],
[
"Chi",
"Ed",
""
],
[
"Schärli",
"Nathanael",
""
],
[
"Zhou",
"Denny",
""
]
] | TITLE: Large Language Models Can Be Easily Distracted by Irrelevant Context
ABSTRACT: Large language models have achieved impressive performance on various natural
language processing tasks. However, so far they have been evaluated primarily
on benchmarks where all information in the input context is relevant for
solving the task. In this work, we investigate the distractibility of large
language models, i.e., how the model problem-solving accuracy can be influenced
by irrelevant context. In particular, we introduce Grade-School Math with
Irrelevant Context (GSM-IC), an arithmetic reasoning dataset with irrelevant
information in the problem description. We use this benchmark to measure the
distractibility of cutting-edge prompting techniques for large language models,
and find that the model performance is dramatically decreased when irrelevant
information is included. We also identify several approaches for mitigating
this deficiency, such as decoding with self-consistency and adding to the
prompt an instruction that tells the language model to ignore the irrelevant
information.
|
2302.04775 | Junkang Wu | Jiawei Chen, Junkang Wu, Jiancan Wu, Sheng Zhou, Xuezhi Cao, Xiangnan
He | Adap-$\tau$: Adaptively Modulating Embedding Magnitude for
Recommendation | null | WWW2023 | 10.1145/3543507.3583363 | null | cs.IR cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent years have witnessed the great successes of embedding-based methods in
recommender systems. Despite their decent performance, we argue one potential
limitation of these methods -- the embedding magnitude has not been explicitly
modulated, which may aggravate popularity bias and training instability,
hindering the model from making a good recommendation. It motivates us to
leverage the embedding normalization in recommendation. By normalizing
user/item embeddings to a specific value, we empirically observe impressive
performance gains (9\% on average) on four real-world datasets. Although
encouraging, we also reveal a serious limitation when applying normalization in
recommendation -- the performance is highly sensitive to the choice of the
temperature $\tau$ which controls the scale of the normalized embeddings.
To fully foster the merits of the normalization while circumvent its
limitation, this work studied on how to adaptively set the proper $\tau$.
Towards this end, we first make a comprehensive analyses of $\tau$ to fully
understand its role on recommendation. We then accordingly develop an adaptive
fine-grained strategy Adap-$\tau$ for the temperature with satisfying four
desirable properties including adaptivity, personalized, efficiency and
model-agnostic. Extensive experiments have been conducted to validate the
effectiveness of the proposal. The code is available at
\url{https://github.com/junkangwu/Adap_tau}.
| [
{
"version": "v1",
"created": "Thu, 9 Feb 2023 17:09:53 GMT"
},
{
"version": "v2",
"created": "Tue, 21 Mar 2023 08:48:20 GMT"
},
{
"version": "v3",
"created": "Tue, 6 Jun 2023 12:52:36 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Chen",
"Jiawei",
""
],
[
"Wu",
"Junkang",
""
],
[
"Wu",
"Jiancan",
""
],
[
"Zhou",
"Sheng",
""
],
[
"Cao",
"Xuezhi",
""
],
[
"He",
"Xiangnan",
""
]
] | TITLE: Adap-$\tau$: Adaptively Modulating Embedding Magnitude for
Recommendation
ABSTRACT: Recent years have witnessed the great successes of embedding-based methods in
recommender systems. Despite their decent performance, we argue one potential
limitation of these methods -- the embedding magnitude has not been explicitly
modulated, which may aggravate popularity bias and training instability,
hindering the model from making a good recommendation. It motivates us to
leverage the embedding normalization in recommendation. By normalizing
user/item embeddings to a specific value, we empirically observe impressive
performance gains (9\% on average) on four real-world datasets. Although
encouraging, we also reveal a serious limitation when applying normalization in
recommendation -- the performance is highly sensitive to the choice of the
temperature $\tau$ which controls the scale of the normalized embeddings.
To fully foster the merits of the normalization while circumvent its
limitation, this work studied on how to adaptively set the proper $\tau$.
Towards this end, we first make a comprehensive analyses of $\tau$ to fully
understand its role on recommendation. We then accordingly develop an adaptive
fine-grained strategy Adap-$\tau$ for the temperature with satisfying four
desirable properties including adaptivity, personalized, efficiency and
model-agnostic. Extensive experiments have been conducted to validate the
effectiveness of the proposal. The code is available at
\url{https://github.com/junkangwu/Adap_tau}.
|
2302.07636 | Timour Igamberdiev | Timour Igamberdiev and Ivan Habernal | DP-BART for Privatized Text Rewriting under Local Differential Privacy | Accepted at ACL Findings 2023 | null | null | null | cs.CR cs.CL | http://creativecommons.org/licenses/by-sa/4.0/ | Privatized text rewriting with local differential privacy (LDP) is a recent
approach that enables sharing of sensitive textual documents while formally
guaranteeing privacy protection to individuals. However, existing systems face
several issues, such as formal mathematical flaws, unrealistic privacy
guarantees, privatization of only individual words, as well as a lack of
transparency and reproducibility. In this paper, we propose a new system
'DP-BART' that largely outperforms existing LDP systems. Our approach uses a
novel clipping method, iterative pruning, and further training of internal
representations which drastically reduces the amount of noise required for DP
guarantees. We run experiments on five textual datasets of varying sizes,
rewriting them at different privacy guarantees and evaluating the rewritten
texts on downstream text classification tasks. Finally, we thoroughly discuss
the privatized text rewriting approach and its limitations, including the
problem of the strict text adjacency constraint in the LDP paradigm that leads
to the high noise requirement.
| [
{
"version": "v1",
"created": "Wed, 15 Feb 2023 13:07:34 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 14:17:46 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Igamberdiev",
"Timour",
""
],
[
"Habernal",
"Ivan",
""
]
] | TITLE: DP-BART for Privatized Text Rewriting under Local Differential Privacy
ABSTRACT: Privatized text rewriting with local differential privacy (LDP) is a recent
approach that enables sharing of sensitive textual documents while formally
guaranteeing privacy protection to individuals. However, existing systems face
several issues, such as formal mathematical flaws, unrealistic privacy
guarantees, privatization of only individual words, as well as a lack of
transparency and reproducibility. In this paper, we propose a new system
'DP-BART' that largely outperforms existing LDP systems. Our approach uses a
novel clipping method, iterative pruning, and further training of internal
representations which drastically reduces the amount of noise required for DP
guarantees. We run experiments on five textual datasets of varying sizes,
rewriting them at different privacy guarantees and evaluating the rewritten
texts on downstream text classification tasks. Finally, we thoroughly discuss
the privatized text rewriting approach and its limitations, including the
problem of the strict text adjacency constraint in the LDP paradigm that leads
to the high noise requirement.
|
2302.11848 | Bo Chen | Bo Chen, Jing Zhang, Fanjin Zhang, Tianyi Han, Yuqing Cheng, Xiaoyan
Li, Yuxiao Dong, and Jie Tang | Web-Scale Academic Name Disambiguation: the WhoIsWho Benchmark,
Leaderboard, and Toolkit | Accepted by KDD 2023 ADS track | null | null | null | cs.IR cs.DL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Name disambiguation -- a fundamental problem in online academic systems -- is
now facing greater challenges with the increasing growth of research papers.
For example, on AMiner, an online academic search platform, about 10% of names
own more than 100 authors. Such real-world challenging cases have not been
effectively addressed by existing researches due to the small-scale or
low-quality datasets that they have used. The development of effective
algorithms is further hampered by a variety of tasks and evaluation protocols
designed on top of diverse datasets. To this end, we present WhoIsWho owning, a
large-scale benchmark with over 1,000,000 papers built using an interactive
annotation process, a regular leaderboard with comprehensive tasks, and an
easy-to-use toolkit encapsulating the entire pipeline as well as the most
powerful features and baseline models for tackling the tasks. Our developed
strong baseline has already been deployed online in the AMiner system to enable
daily arXiv paper assignments. The public leaderboard is available at
http://whoiswho.biendata.xyz/. The toolkit is at
https://github.com/THUDM/WhoIsWho. The online demo of daily arXiv paper
assignments is at https://na-demo.aminer.cn/arxivpaper.
| [
{
"version": "v1",
"created": "Thu, 23 Feb 2023 08:26:35 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 08:41:31 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Chen",
"Bo",
""
],
[
"Zhang",
"Jing",
""
],
[
"Zhang",
"Fanjin",
""
],
[
"Han",
"Tianyi",
""
],
[
"Cheng",
"Yuqing",
""
],
[
"Li",
"Xiaoyan",
""
],
[
"Dong",
"Yuxiao",
""
],
[
"Tang",
"Jie",
""
]
] | TITLE: Web-Scale Academic Name Disambiguation: the WhoIsWho Benchmark,
Leaderboard, and Toolkit
ABSTRACT: Name disambiguation -- a fundamental problem in online academic systems -- is
now facing greater challenges with the increasing growth of research papers.
For example, on AMiner, an online academic search platform, about 10% of names
own more than 100 authors. Such real-world challenging cases have not been
effectively addressed by existing researches due to the small-scale or
low-quality datasets that they have used. The development of effective
algorithms is further hampered by a variety of tasks and evaluation protocols
designed on top of diverse datasets. To this end, we present WhoIsWho owning, a
large-scale benchmark with over 1,000,000 papers built using an interactive
annotation process, a regular leaderboard with comprehensive tasks, and an
easy-to-use toolkit encapsulating the entire pipeline as well as the most
powerful features and baseline models for tackling the tasks. Our developed
strong baseline has already been deployed online in the AMiner system to enable
daily arXiv paper assignments. The public leaderboard is available at
http://whoiswho.biendata.xyz/. The toolkit is at
https://github.com/THUDM/WhoIsWho. The online demo of daily arXiv paper
assignments is at https://na-demo.aminer.cn/arxivpaper.
|
2303.01513 | May Yong | Mahed Abroshan, Michael Burkhart, Oscar Giles, Sam Greenbury, Zoe
Kourtzi, Jack Roberts, Mihaela van der Schaar, Jannetta S Steyn, Alan Wilson,
May Yong | Safe AI for health and beyond -- Monitoring to transform a health
service | 12 pages, 3 figures | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Machine learning techniques are effective for building predictive models
because they identify patterns in large datasets. Development of a model for
complex real-life problems often stop at the point of publication, proof of
concept or when made accessible through some mode of deployment. However, a
model in the medical domain risks becoming obsolete as patient demographics,
systems and clinical practices change. The maintenance and monitoring of
predictive model performance post-publication is crucial to enable their safe
and effective long-term use. We will assess the infrastructure required to
monitor the outputs of a machine learning algorithm, and present two scenarios
with examples of monitoring and updates of models, firstly on a breast cancer
prognosis model trained on public longitudinal data, and secondly on a
neurodegenerative stratification algorithm that is currently being developed
and tested in clinic.
| [
{
"version": "v1",
"created": "Thu, 2 Mar 2023 17:27:45 GMT"
},
{
"version": "v2",
"created": "Mon, 10 Apr 2023 11:51:10 GMT"
},
{
"version": "v3",
"created": "Tue, 6 Jun 2023 12:02:18 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Abroshan",
"Mahed",
""
],
[
"Burkhart",
"Michael",
""
],
[
"Giles",
"Oscar",
""
],
[
"Greenbury",
"Sam",
""
],
[
"Kourtzi",
"Zoe",
""
],
[
"Roberts",
"Jack",
""
],
[
"van der Schaar",
"Mihaela",
""
],
[
"Steyn",
"Jannetta S",
""
],
[
"Wilson",
"Alan",
""
],
[
"Yong",
"May",
""
]
] | TITLE: Safe AI for health and beyond -- Monitoring to transform a health
service
ABSTRACT: Machine learning techniques are effective for building predictive models
because they identify patterns in large datasets. Development of a model for
complex real-life problems often stop at the point of publication, proof of
concept or when made accessible through some mode of deployment. However, a
model in the medical domain risks becoming obsolete as patient demographics,
systems and clinical practices change. The maintenance and monitoring of
predictive model performance post-publication is crucial to enable their safe
and effective long-term use. We will assess the infrastructure required to
monitor the outputs of a machine learning algorithm, and present two scenarios
with examples of monitoring and updates of models, firstly on a breast cancer
prognosis model trained on public longitudinal data, and secondly on a
neurodegenerative stratification algorithm that is currently being developed
and tested in clinic.
|
2303.03592 | Yiwei Lu | Yiwei Lu, Gautam Kamath, Yaoliang Yu | Exploring the Limits of Model-Targeted Indiscriminate Data Poisoning
Attacks | Accepted to ICML 2023 | null | null | null | cs.LG cs.CR | http://creativecommons.org/licenses/by/4.0/ | Indiscriminate data poisoning attacks aim to decrease a model's test accuracy
by injecting a small amount of corrupted training data. Despite significant
interest, existing attacks remain relatively ineffective against modern machine
learning (ML) architectures. In this work, we introduce the notion of model
poisoning reachability as a technical tool to explore the intrinsic limits of
data poisoning attacks towards target parameters (i.e., model-targeted
attacks). We derive an easily computable threshold to establish and quantify a
surprising phase transition phenomenon among popular ML models: data poisoning
attacks can achieve certain target parameters only when the poisoning ratio
exceeds our threshold. Building on existing parameter corruption attacks and
refining the Gradient Canceling attack, we perform extensive experiments to
confirm our theoretical findings, test the predictability of our transition
threshold, and significantly improve existing indiscriminate data poisoning
baselines over a range of datasets and models. Our work highlights the critical
role played by the poisoning ratio, and sheds new insights on existing
empirical results, attacks and mitigation strategies in data poisoning.
| [
{
"version": "v1",
"created": "Tue, 7 Mar 2023 01:55:26 GMT"
},
{
"version": "v2",
"created": "Wed, 8 Mar 2023 01:50:21 GMT"
},
{
"version": "v3",
"created": "Tue, 6 Jun 2023 04:39:06 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Lu",
"Yiwei",
""
],
[
"Kamath",
"Gautam",
""
],
[
"Yu",
"Yaoliang",
""
]
] | TITLE: Exploring the Limits of Model-Targeted Indiscriminate Data Poisoning
Attacks
ABSTRACT: Indiscriminate data poisoning attacks aim to decrease a model's test accuracy
by injecting a small amount of corrupted training data. Despite significant
interest, existing attacks remain relatively ineffective against modern machine
learning (ML) architectures. In this work, we introduce the notion of model
poisoning reachability as a technical tool to explore the intrinsic limits of
data poisoning attacks towards target parameters (i.e., model-targeted
attacks). We derive an easily computable threshold to establish and quantify a
surprising phase transition phenomenon among popular ML models: data poisoning
attacks can achieve certain target parameters only when the poisoning ratio
exceeds our threshold. Building on existing parameter corruption attacks and
refining the Gradient Canceling attack, we perform extensive experiments to
confirm our theoretical findings, test the predictability of our transition
threshold, and significantly improve existing indiscriminate data poisoning
baselines over a range of datasets and models. Our work highlights the critical
role played by the poisoning ratio, and sheds new insights on existing
empirical results, attacks and mitigation strategies in data poisoning.
|
2303.04791 | Arthur Kosmala | Arthur Kosmala, Johannes Gasteiger, Nicholas Gao, Stephan G\"unnemann | Ewald-based Long-Range Message Passing for Molecular Graphs | Published at the 40th International Conference on Machine Learning
(ICML 2023) | null | null | null | cs.LG cond-mat.mtrl-sci physics.chem-ph physics.comp-ph | http://creativecommons.org/licenses/by/4.0/ | Neural architectures that learn potential energy surfaces from molecular data
have undergone fast improvement in recent years. A key driver of this success
is the Message Passing Neural Network (MPNN) paradigm. Its favorable scaling
with system size partly relies upon a spatial distance limit on messages. While
this focus on locality is a useful inductive bias, it also impedes the learning
of long-range interactions such as electrostatics and van der Waals forces. To
address this drawback, we propose Ewald message passing: a nonlocal Fourier
space scheme which limits interactions via a cutoff on frequency instead of
distance, and is theoretically well-founded in the Ewald summation method. It
can serve as an augmentation on top of existing MPNN architectures as it is
computationally inexpensive and agnostic to architectural details. We test the
approach with four baseline models and two datasets containing diverse periodic
(OC20) and aperiodic structures (OE62). We observe robust improvements in
energy mean absolute errors across all models and datasets, averaging 10% on
OC20 and 16% on OE62. Our analysis shows an outsize impact of these
improvements on structures with high long-range contributions to the ground
truth energy.
| [
{
"version": "v1",
"created": "Wed, 8 Mar 2023 18:41:27 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 16:15:34 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Kosmala",
"Arthur",
""
],
[
"Gasteiger",
"Johannes",
""
],
[
"Gao",
"Nicholas",
""
],
[
"Günnemann",
"Stephan",
""
]
] | TITLE: Ewald-based Long-Range Message Passing for Molecular Graphs
ABSTRACT: Neural architectures that learn potential energy surfaces from molecular data
have undergone fast improvement in recent years. A key driver of this success
is the Message Passing Neural Network (MPNN) paradigm. Its favorable scaling
with system size partly relies upon a spatial distance limit on messages. While
this focus on locality is a useful inductive bias, it also impedes the learning
of long-range interactions such as electrostatics and van der Waals forces. To
address this drawback, we propose Ewald message passing: a nonlocal Fourier
space scheme which limits interactions via a cutoff on frequency instead of
distance, and is theoretically well-founded in the Ewald summation method. It
can serve as an augmentation on top of existing MPNN architectures as it is
computationally inexpensive and agnostic to architectural details. We test the
approach with four baseline models and two datasets containing diverse periodic
(OC20) and aperiodic structures (OE62). We observe robust improvements in
energy mean absolute errors across all models and datasets, averaging 10% on
OC20 and 16% on OE62. Our analysis shows an outsize impact of these
improvements on structures with high long-range contributions to the ground
truth energy.
|
2303.05785 | Ho Hin Lee | Ho Hin Lee, Quan Liu, Shunxing Bao, Qi Yang, Xin Yu, Leon Y. Cai,
Thomas Li, Yuankai Huo, Xenofon Koutsoukos, Bennett A. Landman | Scaling Up 3D Kernels with Bayesian Frequency Re-parameterization for
Medical Image Segmentation | Accepted to MICCAI 2023 (top 13.6%), both codes and pretrained models
are available at: https://github.com/MASILab/RepUX-Net | null | null | null | eess.IV cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | With the inspiration of vision transformers, the concept of depth-wise
convolution revisits to provide a large Effective Receptive Field (ERF) using
Large Kernel (LK) sizes for medical image segmentation. However, the
segmentation performance might be saturated and even degraded as the kernel
sizes scaled up (e.g., $21\times 21\times 21$) in a Convolutional Neural
Network (CNN). We hypothesize that convolution with LK sizes is limited to
maintain an optimal convergence for locality learning. While Structural
Re-parameterization (SR) enhances the local convergence with small kernels in
parallel, optimal small kernel branches may hinder the computational efficiency
for training. In this work, we propose RepUX-Net, a pure CNN architecture with
a simple large kernel block design, which competes favorably with current
network state-of-the-art (SOTA) (e.g., 3D UX-Net, SwinUNETR) using 6
challenging public datasets. We derive an equivalency between kernel
re-parameterization and the branch-wise variation in kernel convergence.
Inspired by the spatial frequency in the human visual system, we extend to vary
the kernel convergence into element-wise setting and model the spatial
frequency as a Bayesian prior to re-parameterize convolutional weights during
training. Specifically, a reciprocal function is leveraged to estimate a
frequency-weighted value, which rescales the corresponding kernel element for
stochastic gradient descent. From the experimental results, RepUX-Net
consistently outperforms 3D SOTA benchmarks with internal validation (FLARE:
0.929 to 0.944), external validation (MSD: 0.901 to 0.932, KiTS: 0.815 to
0.847, LiTS: 0.933 to 0.949, TCIA: 0.736 to 0.779) and transfer learning (AMOS:
0.880 to 0.911) scenarios in Dice Score.
| [
{
"version": "v1",
"created": "Fri, 10 Mar 2023 08:38:34 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 03:05:07 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Lee",
"Ho Hin",
""
],
[
"Liu",
"Quan",
""
],
[
"Bao",
"Shunxing",
""
],
[
"Yang",
"Qi",
""
],
[
"Yu",
"Xin",
""
],
[
"Cai",
"Leon Y.",
""
],
[
"Li",
"Thomas",
""
],
[
"Huo",
"Yuankai",
""
],
[
"Koutsoukos",
"Xenofon",
""
],
[
"Landman",
"Bennett A.",
""
]
] | TITLE: Scaling Up 3D Kernels with Bayesian Frequency Re-parameterization for
Medical Image Segmentation
ABSTRACT: With the inspiration of vision transformers, the concept of depth-wise
convolution revisits to provide a large Effective Receptive Field (ERF) using
Large Kernel (LK) sizes for medical image segmentation. However, the
segmentation performance might be saturated and even degraded as the kernel
sizes scaled up (e.g., $21\times 21\times 21$) in a Convolutional Neural
Network (CNN). We hypothesize that convolution with LK sizes is limited to
maintain an optimal convergence for locality learning. While Structural
Re-parameterization (SR) enhances the local convergence with small kernels in
parallel, optimal small kernel branches may hinder the computational efficiency
for training. In this work, we propose RepUX-Net, a pure CNN architecture with
a simple large kernel block design, which competes favorably with current
network state-of-the-art (SOTA) (e.g., 3D UX-Net, SwinUNETR) using 6
challenging public datasets. We derive an equivalency between kernel
re-parameterization and the branch-wise variation in kernel convergence.
Inspired by the spatial frequency in the human visual system, we extend to vary
the kernel convergence into element-wise setting and model the spatial
frequency as a Bayesian prior to re-parameterize convolutional weights during
training. Specifically, a reciprocal function is leveraged to estimate a
frequency-weighted value, which rescales the corresponding kernel element for
stochastic gradient descent. From the experimental results, RepUX-Net
consistently outperforms 3D SOTA benchmarks with internal validation (FLARE:
0.929 to 0.944), external validation (MSD: 0.901 to 0.932, KiTS: 0.815 to
0.847, LiTS: 0.933 to 0.949, TCIA: 0.736 to 0.779) and transfer learning (AMOS:
0.880 to 0.911) scenarios in Dice Score.
|
2303.10909 | Jeongwhan Choi | Jeongwhan Choi, Noseong Park | Graph Neural Rough Differential Equations for Traffic Forecasting | Accepted to ACM Transactions on Intelligent Systems and Technology
(ACM TIST). arXiv admin note: substantial text overlap with arXiv:2112.03558 | null | null | null | cs.LG cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Traffic forecasting is one of the most popular spatio-temporal tasks in the
field of machine learning. A prevalent approach in the field is to combine
graph convolutional networks and recurrent neural networks for the
spatio-temporal processing. There has been fierce competition and many novel
methods have been proposed. In this paper, we present the method of
spatio-temporal graph neural rough differential equation (STG-NRDE). Neural
rough differential equations (NRDEs) are a breakthrough concept for processing
time-series data. Their main concept is to use the log-signature transform to
convert a time-series sample into a relatively shorter series of feature
vectors. We extend the concept and design two NRDEs: one for the temporal
processing and the other for the spatial processing. After that, we combine
them into a single framework. We conduct experiments with 6 benchmark datasets
and 27 baselines. STG-NRDE shows the best accuracy in all cases, outperforming
all those 27 baselines by non-trivial margins.
| [
{
"version": "v1",
"created": "Mon, 20 Mar 2023 06:57:10 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 08:11:51 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Choi",
"Jeongwhan",
""
],
[
"Park",
"Noseong",
""
]
] | TITLE: Graph Neural Rough Differential Equations for Traffic Forecasting
ABSTRACT: Traffic forecasting is one of the most popular spatio-temporal tasks in the
field of machine learning. A prevalent approach in the field is to combine
graph convolutional networks and recurrent neural networks for the
spatio-temporal processing. There has been fierce competition and many novel
methods have been proposed. In this paper, we present the method of
spatio-temporal graph neural rough differential equation (STG-NRDE). Neural
rough differential equations (NRDEs) are a breakthrough concept for processing
time-series data. Their main concept is to use the log-signature transform to
convert a time-series sample into a relatively shorter series of feature
vectors. We extend the concept and design two NRDEs: one for the temporal
processing and the other for the spatial processing. After that, we combine
them into a single framework. We conduct experiments with 6 benchmark datasets
and 27 baselines. STG-NRDE shows the best accuracy in all cases, outperforming
all those 27 baselines by non-trivial margins.
|
2303.13190 | Weixiao Liu | Weixiao Liu, Yuwei Wu, Sipu Ruan, Gregory S. Chirikjian | Marching-Primitives: Shape Abstraction from Signed Distance Function | Accepted to CVPR2023 Highlight | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Representing complex objects with basic geometric primitives has long been a
topic in computer vision. Primitive-based representations have the merits of
compactness and computational efficiency in higher-level tasks such as physics
simulation, collision checking, and robotic manipulation. Unlike previous works
which extract polygonal meshes from a signed distance function (SDF), in this
paper, we present a novel method, named Marching-Primitives, to obtain a
primitive-based abstraction directly from an SDF. Our method grows geometric
primitives (such as superquadrics) iteratively by analyzing the connectivity of
voxels while marching at different levels of signed distance. For each valid
connected volume of interest, we march on the scope of voxels from which a
primitive is able to be extracted in a probabilistic sense and simultaneously
solve for the parameters of the primitive to capture the underlying local
geometry. We evaluate the performance of our method on both synthetic and
real-world datasets. The results show that the proposed method outperforms the
state-of-the-art in terms of accuracy, and is directly generalizable among
different categories and scales. The code is open-sourced at
https://github.com/ChirikjianLab/Marching-Primitives.git.
| [
{
"version": "v1",
"created": "Thu, 23 Mar 2023 11:42:35 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 03:21:47 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Liu",
"Weixiao",
""
],
[
"Wu",
"Yuwei",
""
],
[
"Ruan",
"Sipu",
""
],
[
"Chirikjian",
"Gregory S.",
""
]
] | TITLE: Marching-Primitives: Shape Abstraction from Signed Distance Function
ABSTRACT: Representing complex objects with basic geometric primitives has long been a
topic in computer vision. Primitive-based representations have the merits of
compactness and computational efficiency in higher-level tasks such as physics
simulation, collision checking, and robotic manipulation. Unlike previous works
which extract polygonal meshes from a signed distance function (SDF), in this
paper, we present a novel method, named Marching-Primitives, to obtain a
primitive-based abstraction directly from an SDF. Our method grows geometric
primitives (such as superquadrics) iteratively by analyzing the connectivity of
voxels while marching at different levels of signed distance. For each valid
connected volume of interest, we march on the scope of voxels from which a
primitive is able to be extracted in a probabilistic sense and simultaneously
solve for the parameters of the primitive to capture the underlying local
geometry. We evaluate the performance of our method on both synthetic and
real-world datasets. The results show that the proposed method outperforms the
state-of-the-art in terms of accuracy, and is directly generalizable among
different categories and scales. The code is open-sourced at
https://github.com/ChirikjianLab/Marching-Primitives.git.
|
2303.13434 | Jinjing Zhu | Jinjing Zhu, Haotian Bai, Lin Wang | Patch-Mix Transformer for Unsupervised Domain Adaptation: A Game
Perspective | Accepted by CVPR 2023 (Highlight) | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Endeavors have been recently made to leverage the vision transformer (ViT)
for the challenging unsupervised domain adaptation (UDA) task. They typically
adopt the cross-attention in ViT for direct domain alignment. However, as the
performance of cross-attention highly relies on the quality of pseudo labels
for targeted samples, it becomes less effective when the domain gap becomes
large. We solve this problem from a game theory's perspective with the proposed
model dubbed as PMTrans, which bridges source and target domains with an
intermediate domain. Specifically, we propose a novel ViT-based module called
PatchMix that effectively builds up the intermediate domain, i.e., probability
distribution, by learning to sample patches from both domains based on the
game-theoretical models. This way, it learns to mix the patches from the source
and target domains to maximize the cross entropy (CE), while exploiting two
semi-supervised mixup losses in the feature and label spaces to minimize it. As
such, we interpret the process of UDA as a min-max CE game with three players,
including the feature extractor, classifier, and PatchMix, to find the Nash
Equilibria. Moreover, we leverage attention maps from ViT to re-weight the
label of each patch by its importance, making it possible to obtain more
domain-discriminative feature representations. We conduct extensive experiments
on four benchmark datasets, and the results show that PMTrans significantly
surpasses the ViT-based and CNN-based SoTA methods by +3.6% on Office-Home,
+1.4% on Office-31, and +17.7% on DomainNet, respectively.
| [
{
"version": "v1",
"created": "Thu, 23 Mar 2023 16:56:01 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 04:43:11 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Zhu",
"Jinjing",
""
],
[
"Bai",
"Haotian",
""
],
[
"Wang",
"Lin",
""
]
] | TITLE: Patch-Mix Transformer for Unsupervised Domain Adaptation: A Game
Perspective
ABSTRACT: Endeavors have been recently made to leverage the vision transformer (ViT)
for the challenging unsupervised domain adaptation (UDA) task. They typically
adopt the cross-attention in ViT for direct domain alignment. However, as the
performance of cross-attention highly relies on the quality of pseudo labels
for targeted samples, it becomes less effective when the domain gap becomes
large. We solve this problem from a game theory's perspective with the proposed
model dubbed as PMTrans, which bridges source and target domains with an
intermediate domain. Specifically, we propose a novel ViT-based module called
PatchMix that effectively builds up the intermediate domain, i.e., probability
distribution, by learning to sample patches from both domains based on the
game-theoretical models. This way, it learns to mix the patches from the source
and target domains to maximize the cross entropy (CE), while exploiting two
semi-supervised mixup losses in the feature and label spaces to minimize it. As
such, we interpret the process of UDA as a min-max CE game with three players,
including the feature extractor, classifier, and PatchMix, to find the Nash
Equilibria. Moreover, we leverage attention maps from ViT to re-weight the
label of each patch by its importance, making it possible to obtain more
domain-discriminative feature representations. We conduct extensive experiments
on four benchmark datasets, and the results show that PMTrans significantly
surpasses the ViT-based and CNN-based SoTA methods by +3.6% on Office-Home,
+1.4% on Office-31, and +17.7% on DomainNet, respectively.
|
2303.14829 | Asmar Nadeem | Asmar Nadeem, Adrian Hilton, Robert Dawes, Graham Thomas, Armin
Mustafa | SEM-POS: Grammatically and Semantically Correct Video Captioning | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Generating grammatically and semantically correct captions in video
captioning is a challenging task. The captions generated from the existing
methods are either word-by-word that do not align with grammatical structure or
miss key information from the input videos. To address these issues, we
introduce a novel global-local fusion network, with a Global-Local Fusion Block
(GLFB) that encodes and fuses features from different parts of speech (POS)
components with visual-spatial features. We use novel combinations of different
POS components - 'determinant + subject', 'auxiliary verb', 'verb', and
'determinant + object' for supervision of the POS blocks - Det + Subject, Aux
Verb, Verb, and Det + Object respectively. The novel global-local fusion
network together with POS blocks helps align the visual features with language
description to generate grammatically and semantically correct captions.
Extensive qualitative and quantitative experiments on benchmark MSVD and MSRVTT
datasets demonstrate that the proposed approach generates more grammatically
and semantically correct captions compared to the existing methods, achieving
the new state-of-the-art. Ablations on the POS blocks and the GLFB demonstrate
the impact of the contributions on the proposed method.
| [
{
"version": "v1",
"created": "Sun, 26 Mar 2023 21:42:40 GMT"
},
{
"version": "v2",
"created": "Tue, 4 Apr 2023 13:51:34 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Nadeem",
"Asmar",
""
],
[
"Hilton",
"Adrian",
""
],
[
"Dawes",
"Robert",
""
],
[
"Thomas",
"Graham",
""
],
[
"Mustafa",
"Armin",
""
]
] | TITLE: SEM-POS: Grammatically and Semantically Correct Video Captioning
ABSTRACT: Generating grammatically and semantically correct captions in video
captioning is a challenging task. The captions generated from the existing
methods are either word-by-word that do not align with grammatical structure or
miss key information from the input videos. To address these issues, we
introduce a novel global-local fusion network, with a Global-Local Fusion Block
(GLFB) that encodes and fuses features from different parts of speech (POS)
components with visual-spatial features. We use novel combinations of different
POS components - 'determinant + subject', 'auxiliary verb', 'verb', and
'determinant + object' for supervision of the POS blocks - Det + Subject, Aux
Verb, Verb, and Det + Object respectively. The novel global-local fusion
network together with POS blocks helps align the visual features with language
description to generate grammatically and semantically correct captions.
Extensive qualitative and quantitative experiments on benchmark MSVD and MSRVTT
datasets demonstrate that the proposed approach generates more grammatically
and semantically correct captions compared to the existing methods, achieving
the new state-of-the-art. Ablations on the POS blocks and the GLFB demonstrate
the impact of the contributions on the proposed method.
|
2303.17612 | Daniel Campos | Daniel Campos, Alexandre Marques, Mark Kurtz, and ChengXiang Zhai | oBERTa: Improving Sparse Transfer Learning via improved initialization,
distillation, and pruning regimes | SustaiNLP2023 @ ACL 2023,9 pages, 2 figures, 45 tables | null | null | null | cs.CL cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | In this paper, we introduce the range of oBERTa language models, an
easy-to-use set of language models which allows Natural Language Processing
(NLP) practitioners to obtain between 3.8 and 24.3 times faster models without
expertise in model compression. Specifically, oBERTa extends existing work on
pruning, knowledge distillation, and quantization and leverages frozen
embeddings improves distillation and model initialization to deliver higher
accuracy on a broad range of transfer tasks. In generating oBERTa, we explore
how the highly optimized RoBERTa differs from the BERT for pruning during
pre-training and finetuning. We find it less amenable to compression during
fine-tuning. We explore the use of oBERTa on seven representative NLP tasks and
find that the improved compression techniques allow a pruned oBERTa model to
match the performance of BERTbase and exceed the performance of Prune OFA Large
on the SQUAD V1.1 Question Answering dataset, despite being 8x and 2x,
respectively faster in inference. We release our code, training regimes, and
associated model for broad usage to encourage usage and experimentation
| [
{
"version": "v1",
"created": "Thu, 30 Mar 2023 01:37:19 GMT"
},
{
"version": "v2",
"created": "Tue, 4 Apr 2023 15:51:59 GMT"
},
{
"version": "v3",
"created": "Tue, 6 Jun 2023 16:30:09 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Campos",
"Daniel",
""
],
[
"Marques",
"Alexandre",
""
],
[
"Kurtz",
"Mark",
""
],
[
"Zhai",
"ChengXiang",
""
]
] | TITLE: oBERTa: Improving Sparse Transfer Learning via improved initialization,
distillation, and pruning regimes
ABSTRACT: In this paper, we introduce the range of oBERTa language models, an
easy-to-use set of language models which allows Natural Language Processing
(NLP) practitioners to obtain between 3.8 and 24.3 times faster models without
expertise in model compression. Specifically, oBERTa extends existing work on
pruning, knowledge distillation, and quantization and leverages frozen
embeddings improves distillation and model initialization to deliver higher
accuracy on a broad range of transfer tasks. In generating oBERTa, we explore
how the highly optimized RoBERTa differs from the BERT for pruning during
pre-training and finetuning. We find it less amenable to compression during
fine-tuning. We explore the use of oBERTa on seven representative NLP tasks and
find that the improved compression techniques allow a pruned oBERTa model to
match the performance of BERTbase and exceed the performance of Prune OFA Large
on the SQUAD V1.1 Question Answering dataset, despite being 8x and 2x,
respectively faster in inference. We release our code, training regimes, and
associated model for broad usage to encourage usage and experimentation
|
2304.04874 | Eslam Bakr | Eslam Mohamed Bakr, Pengzhan Sun, Li Erran Li, Mohamed Elhoseiny | ImageCaptioner$^2$: Image Captioner for Image Captioning Bias
Amplification Assessment | null | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Most pre-trained learning systems are known to suffer from bias, which
typically emerges from the data, the model, or both. Measuring and quantifying
bias and its sources is a challenging task and has been extensively studied in
image captioning. Despite the significant effort in this direction, we observed
that existing metrics lack consistency in the inclusion of the visual signal.
In this paper, we introduce a new bias assessment metric, dubbed
$ImageCaptioner^2$, for image captioning. Instead of measuring the absolute
bias in the model or the data, $ImageCaptioner^2$ pay more attention to the
bias introduced by the model w.r.t the data bias, termed bias amplification.
Unlike the existing methods, which only evaluate the image captioning
algorithms based on the generated captions only, $ImageCaptioner^2$
incorporates the image while measuring the bias. In addition, we design a
formulation for measuring the bias of generated captions as prompt-based image
captioning instead of using language classifiers. Finally, we apply our
$ImageCaptioner^2$ metric across 11 different image captioning architectures on
three different datasets, i.e., MS-COCO caption dataset, Artemis V1, and
Artemis V2, and on three different protected attributes, i.e., gender, race,
and emotions. Consequently, we verify the effectiveness of our
$ImageCaptioner^2$ metric by proposing AnonymousBench, which is a novel human
evaluation paradigm for bias metrics. Our metric shows significant superiority
over the recent bias metric; LIC, in terms of human alignment, where the
correlation scores are 80% and 54% for our metric and LIC, respectively. The
code is available at https://eslambakr.github.io/imagecaptioner2.github.io/.
| [
{
"version": "v1",
"created": "Mon, 10 Apr 2023 21:40:46 GMT"
},
{
"version": "v2",
"created": "Mon, 5 Jun 2023 22:06:07 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Bakr",
"Eslam Mohamed",
""
],
[
"Sun",
"Pengzhan",
""
],
[
"Li",
"Li Erran",
""
],
[
"Elhoseiny",
"Mohamed",
""
]
] | TITLE: ImageCaptioner$^2$: Image Captioner for Image Captioning Bias
Amplification Assessment
ABSTRACT: Most pre-trained learning systems are known to suffer from bias, which
typically emerges from the data, the model, or both. Measuring and quantifying
bias and its sources is a challenging task and has been extensively studied in
image captioning. Despite the significant effort in this direction, we observed
that existing metrics lack consistency in the inclusion of the visual signal.
In this paper, we introduce a new bias assessment metric, dubbed
$ImageCaptioner^2$, for image captioning. Instead of measuring the absolute
bias in the model or the data, $ImageCaptioner^2$ pay more attention to the
bias introduced by the model w.r.t the data bias, termed bias amplification.
Unlike the existing methods, which only evaluate the image captioning
algorithms based on the generated captions only, $ImageCaptioner^2$
incorporates the image while measuring the bias. In addition, we design a
formulation for measuring the bias of generated captions as prompt-based image
captioning instead of using language classifiers. Finally, we apply our
$ImageCaptioner^2$ metric across 11 different image captioning architectures on
three different datasets, i.e., MS-COCO caption dataset, Artemis V1, and
Artemis V2, and on three different protected attributes, i.e., gender, race,
and emotions. Consequently, we verify the effectiveness of our
$ImageCaptioner^2$ metric by proposing AnonymousBench, which is a novel human
evaluation paradigm for bias metrics. Our metric shows significant superiority
over the recent bias metric; LIC, in terms of human alignment, where the
correlation scores are 80% and 54% for our metric and LIC, respectively. The
code is available at https://eslambakr.github.io/imagecaptioner2.github.io/.
|
2304.06447 | Yihao Ding | Yihao Ding, Siwen Luo, Hyunsuk Chung, Soyeon Caren Han | PDFVQA: A New Dataset for Real-World VQA on PDF Documents | Accepted by ECML-PKDD 2023 | null | null | null | cs.CV cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Document-based Visual Question Answering examines the document understanding
of document images in conditions of natural language questions. We proposed a
new document-based VQA dataset, PDF-VQA, to comprehensively examine the
document understanding from various aspects, including document element
recognition, document layout structural understanding as well as contextual
understanding and key information extraction. Our PDF-VQA dataset extends the
current scale of document understanding that limits on the single document page
to the new scale that asks questions over the full document of multiple pages.
We also propose a new graph-based VQA model that explicitly integrates the
spatial and hierarchically structural relationships between different document
elements to boost the document structural understanding. The performances are
compared with several baselines over different question types and
tasks\footnote{The full dataset will be released after paper acceptance.
| [
{
"version": "v1",
"created": "Thu, 13 Apr 2023 12:28:14 GMT"
},
{
"version": "v2",
"created": "Fri, 14 Apr 2023 02:58:00 GMT"
},
{
"version": "v3",
"created": "Wed, 19 Apr 2023 14:10:08 GMT"
},
{
"version": "v4",
"created": "Mon, 24 Apr 2023 01:46:17 GMT"
},
{
"version": "v5",
"created": "Tue, 6 Jun 2023 02:26:42 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Ding",
"Yihao",
""
],
[
"Luo",
"Siwen",
""
],
[
"Chung",
"Hyunsuk",
""
],
[
"Han",
"Soyeon Caren",
""
]
] | TITLE: PDFVQA: A New Dataset for Real-World VQA on PDF Documents
ABSTRACT: Document-based Visual Question Answering examines the document understanding
of document images in conditions of natural language questions. We proposed a
new document-based VQA dataset, PDF-VQA, to comprehensively examine the
document understanding from various aspects, including document element
recognition, document layout structural understanding as well as contextual
understanding and key information extraction. Our PDF-VQA dataset extends the
current scale of document understanding that limits on the single document page
to the new scale that asks questions over the full document of multiple pages.
We also propose a new graph-based VQA model that explicitly integrates the
spatial and hierarchically structural relationships between different document
elements to boost the document structural understanding. The performances are
compared with several baselines over different question types and
tasks\footnote{The full dataset will be released after paper acceptance.
|
2304.09836 | \'Etienne Marcotte | \'Etienne Marcotte, Valentina Zantedeschi, Alexandre Drouin, Nicolas
Chapados | Regions of Reliability in the Evaluation of Multivariate Probabilistic
Forecasts | 47 pages, 37 figures, camera-ready version, Fortieth International
Conference on Machine Learning (ICML 2023) | null | null | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multivariate probabilistic time series forecasts are commonly evaluated via
proper scoring rules, i.e., functions that are minimal in expectation for the
ground-truth distribution. However, this property is not sufficient to
guarantee good discrimination in the non-asymptotic regime. In this paper, we
provide the first systematic finite-sample study of proper scoring rules for
time-series forecasting evaluation. Through a power analysis, we identify the
"region of reliability" of a scoring rule, i.e., the set of practical
conditions where it can be relied on to identify forecasting errors. We carry
out our analysis on a comprehensive synthetic benchmark, specifically designed
to test several key discrepancies between ground-truth and forecast
distributions, and we gauge the generalizability of our findings to real-world
tasks with an application to an electricity production problem. Our results
reveal critical shortcomings in the evaluation of multivariate probabilistic
forecasts as commonly performed in the literature.
| [
{
"version": "v1",
"created": "Wed, 19 Apr 2023 17:38:42 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 15:39:51 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Marcotte",
"Étienne",
""
],
[
"Zantedeschi",
"Valentina",
""
],
[
"Drouin",
"Alexandre",
""
],
[
"Chapados",
"Nicolas",
""
]
] | TITLE: Regions of Reliability in the Evaluation of Multivariate Probabilistic
Forecasts
ABSTRACT: Multivariate probabilistic time series forecasts are commonly evaluated via
proper scoring rules, i.e., functions that are minimal in expectation for the
ground-truth distribution. However, this property is not sufficient to
guarantee good discrimination in the non-asymptotic regime. In this paper, we
provide the first systematic finite-sample study of proper scoring rules for
time-series forecasting evaluation. Through a power analysis, we identify the
"region of reliability" of a scoring rule, i.e., the set of practical
conditions where it can be relied on to identify forecasting errors. We carry
out our analysis on a comprehensive synthetic benchmark, specifically designed
to test several key discrepancies between ground-truth and forecast
distributions, and we gauge the generalizability of our findings to real-world
tasks with an application to an electricity production problem. Our results
reveal critical shortcomings in the evaluation of multivariate probabilistic
forecasts as commonly performed in the literature.
|
2304.11966 | Wenwen Yu | Wenwen Yu, Mingyu Liu, Mingrui Chen, Ning Lu, Yinlong Wen, Yuliang
Liu, Dimosthenis Karatzas, Xiang Bai | ICDAR 2023 Competition on Reading the Seal Title | ICDAR2023 Competition on ReST report (To be appear in ICDAR 2023) | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Reading seal title text is a challenging task due to the variable shapes of
seals, curved text, background noise, and overlapped text. However, this
important element is commonly found in official and financial scenarios, and
has not received the attention it deserves in the field of OCR technology. To
promote research in this area, we organized ICDAR 2023 competition on reading
the seal title (ReST), which included two tasks: seal title text detection
(Task 1) and end-to-end seal title recognition (Task 2). We constructed a
dataset of 10,000 real seal data, covering the most common classes of seals,
and labeled all seal title texts with text polygons and text contents. The
competition opened on 30th December, 2022 and closed on 20th March, 2023. The
competition attracted 53 participants from academia and industry including 28
submissions for Task 1 and 25 submissions for Task 2, which demonstrated
significant interest in this challenging task. In this report, we present an
overview of the competition, including the organization, challenges, and
results. We describe the dataset and tasks, and summarize the submissions and
evaluation results. The results show that significant progress has been made in
the field of seal title text reading, and we hope that this competition will
inspire further research and development in this important area of OCR
technology.
| [
{
"version": "v1",
"created": "Mon, 24 Apr 2023 10:01:41 GMT"
},
{
"version": "v2",
"created": "Mon, 5 Jun 2023 21:56:29 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Yu",
"Wenwen",
""
],
[
"Liu",
"Mingyu",
""
],
[
"Chen",
"Mingrui",
""
],
[
"Lu",
"Ning",
""
],
[
"Wen",
"Yinlong",
""
],
[
"Liu",
"Yuliang",
""
],
[
"Karatzas",
"Dimosthenis",
""
],
[
"Bai",
"Xiang",
""
]
] | TITLE: ICDAR 2023 Competition on Reading the Seal Title
ABSTRACT: Reading seal title text is a challenging task due to the variable shapes of
seals, curved text, background noise, and overlapped text. However, this
important element is commonly found in official and financial scenarios, and
has not received the attention it deserves in the field of OCR technology. To
promote research in this area, we organized ICDAR 2023 competition on reading
the seal title (ReST), which included two tasks: seal title text detection
(Task 1) and end-to-end seal title recognition (Task 2). We constructed a
dataset of 10,000 real seal data, covering the most common classes of seals,
and labeled all seal title texts with text polygons and text contents. The
competition opened on 30th December, 2022 and closed on 20th March, 2023. The
competition attracted 53 participants from academia and industry including 28
submissions for Task 1 and 25 submissions for Task 2, which demonstrated
significant interest in this challenging task. In this report, we present an
overview of the competition, including the organization, challenges, and
results. We describe the dataset and tasks, and summarize the submissions and
evaluation results. The results show that significant progress has been made in
the field of seal title text reading, and we hope that this competition will
inspire further research and development in this important area of OCR
technology.
|
2304.13416 | Shitong Shao | Shitong Shao, Xiaohan Yuan, Zhen Huang, Ziming Qiu, Shuai Wang and
Kevin Zhou | DiffuseExpand: Expanding dataset for 2D medical image segmentation using
diffusion models | Accepted by IJCAI workshop (1st International Workshop on
Generalizing from Limited Resources in the Open World). pre-version was
rejected by MICCAI | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Dataset expansion can effectively alleviate the problem of data scarcity for
medical image segmentation, due to privacy concerns and labeling difficulties.
However, existing expansion algorithms still face great challenges due to their
inability of guaranteeing the diversity of synthesized images with paired
segmentation masks. In recent years, Diffusion Probabilistic Models (DPMs) have
shown powerful image synthesis performance, even better than Generative
Adversarial Networks. Based on this insight, we propose an approach called
DiffuseExpand for expanding datasets for 2D medical image segmentation using
DPM, which first samples a variety of masks from Gaussian noise to ensure the
diversity, and then synthesizes images to ensure the alignment of images and
masks. After that, DiffuseExpand chooses high-quality samples to further
enhance the effectiveness of data expansion. Our comparison and ablation
experiments on COVID-19 and CGMH Pelvis datasets demonstrate the effectiveness
of DiffuseExpand. Our code is released at
https://github.com/shaoshitong/DiffuseExpand.
| [
{
"version": "v1",
"created": "Wed, 26 Apr 2023 09:55:12 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 09:44:19 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Shao",
"Shitong",
""
],
[
"Yuan",
"Xiaohan",
""
],
[
"Huang",
"Zhen",
""
],
[
"Qiu",
"Ziming",
""
],
[
"Wang",
"Shuai",
""
],
[
"Zhou",
"Kevin",
""
]
] | TITLE: DiffuseExpand: Expanding dataset for 2D medical image segmentation using
diffusion models
ABSTRACT: Dataset expansion can effectively alleviate the problem of data scarcity for
medical image segmentation, due to privacy concerns and labeling difficulties.
However, existing expansion algorithms still face great challenges due to their
inability of guaranteeing the diversity of synthesized images with paired
segmentation masks. In recent years, Diffusion Probabilistic Models (DPMs) have
shown powerful image synthesis performance, even better than Generative
Adversarial Networks. Based on this insight, we propose an approach called
DiffuseExpand for expanding datasets for 2D medical image segmentation using
DPM, which first samples a variety of masks from Gaussian noise to ensure the
diversity, and then synthesizes images to ensure the alignment of images and
masks. After that, DiffuseExpand chooses high-quality samples to further
enhance the effectiveness of data expansion. Our comparison and ablation
experiments on COVID-19 and CGMH Pelvis datasets demonstrate the effectiveness
of DiffuseExpand. Our code is released at
https://github.com/shaoshitong/DiffuseExpand.
|
2304.14953 | Micha{\l} Turski | Micha{\l} Turski, Tomasz Stanis{\l}awek, Karol Kaczmarek, Pawe{\l}
Dyda, and Filip Grali\'nski | CCpdf: Building a High Quality Corpus for Visually Rich Documents from
Web Crawl Data | Accepted at ICDAR 2023 | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In recent years, the field of document understanding has progressed a lot. A
significant part of this progress has been possible thanks to the use of
language models pretrained on large amounts of documents. However, pretraining
corpora used in the domain of document understanding are single domain,
monolingual, or nonpublic. Our goal in this paper is to propose an efficient
pipeline for creating a big-scale, diverse, multilingual corpus of PDF files
from all over the Internet using Common Crawl, as PDF files are the most
canonical types of documents as considered in document understanding. We
analysed extensively all of the steps of the pipeline and proposed a solution
which is a trade-off between data quality and processing time. We also share a
CCpdf corpus in a form or an index of PDF files along with a script for
downloading them, which produces a collection useful for language model
pretraining. The dataset and tools published with this paper offer researchers
the opportunity to develop even better multilingual language models.
| [
{
"version": "v1",
"created": "Fri, 28 Apr 2023 16:12:18 GMT"
},
{
"version": "v2",
"created": "Tue, 6 Jun 2023 07:35:17 GMT"
}
] | 2023-06-07T00:00:00 | [
[
"Turski",
"Michał",
""
],
[
"Stanisławek",
"Tomasz",
""
],
[
"Kaczmarek",
"Karol",
""
],
[
"Dyda",
"Paweł",
""
],
[
"Graliński",
"Filip",
""
]
] | TITLE: CCpdf: Building a High Quality Corpus for Visually Rich Documents from
Web Crawl Data
ABSTRACT: In recent years, the field of document understanding has progressed a lot. A
significant part of this progress has been possible thanks to the use of
language models pretrained on large amounts of documents. However, pretraining
corpora used in the domain of document understanding are single domain,
monolingual, or nonpublic. Our goal in this paper is to propose an efficient
pipeline for creating a big-scale, diverse, multilingual corpus of PDF files
from all over the Internet using Common Crawl, as PDF files are the most
canonical types of documents as considered in document understanding. We
analysed extensively all of the steps of the pipeline and proposed a solution
which is a trade-off between data quality and processing time. We also share a
CCpdf corpus in a form or an index of PDF files along with a script for
downloading them, which produces a collection useful for language model
pretraining. The dataset and tools published with this paper offer researchers
the opportunity to develop even better multilingual language models.
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.