Search is not available for this dataset
id
string
submitter
string
authors
string
title
string
comments
string
journal-ref
string
doi
string
report-no
string
categories
string
license
string
abstract
string
versions
list
update_date
timestamp[s]
authors_parsed
list
prompt
string
2504.05855
Zhang Dong
Xingzu Liu, Songhang deng, Mingbang Wang, Zhang Dong, Le Dai, Jiyuan Li, Ruilin Nong
Enhancing Coreference Resolution with Pretrained Language Models: Bridging the Gap Between Syntax and Semantics
acl submission
null
null
null
cs.CL cs.AI
http://creativecommons.org/licenses/by-sa/4.0/
Large language models have made significant advancements in various natural language processing tasks, including coreference resolution. However, traditional methods often fall short in effectively distinguishing referential relationships due to a lack of integration between syntactic and semantic information. This s...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 09:33:09 GMT" } ]
2025-04-09T00:00:00
[ [ "Liu", "Xingzu", "" ], [ "deng", "Songhang", "" ], [ "Wang", "Mingbang", "" ], [ "Dong", "Zhang", "" ], [ "Dai", "Le", "" ], [ "Li", "Jiyuan", "" ], [ "Nong", "Ruilin", "" ] ]
TITLE: Enhancing Coreference Resolution with Pretrained Language Models: Bridging the Gap Between Syntax and Semantics ABSTRACT: Large language models have made significant advancements in various natural language processing tasks, including coreference resolution. However, traditional methods often fall short in...
2504.05866
Sofia Della Penna
Sofia Della Penna, Roberto Natella, Vittorio Orbinato, Lorenzo Parracino, Luciano Pianese
CTI-HAL: A Human-Annotated Dataset for Cyber Threat Intelligence Analysis
Accepted for publication in the Workshop on Attackers and Cybercrime Operations (WACCO 2025), co-located with IEEE European Symposium on Security and Privacy 2025
null
null
null
cs.CR
http://creativecommons.org/licenses/by/4.0/
Organizations are increasingly targeted by Advanced Persistent Threats (APTs), which involve complex, multi-stage tactics and diverse techniques. Cyber Threat Intelligence (CTI) sources, such as incident reports and security blogs, provide valuable insights, but are often unstructured and in natural language, making ...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 09:47:15 GMT" } ]
2025-04-09T00:00:00
[ [ "Della Penna", "Sofia", "" ], [ "Natella", "Roberto", "" ], [ "Orbinato", "Vittorio", "" ], [ "Parracino", "Lorenzo", "" ], [ "Pianese", "Luciano", "" ] ]
TITLE: CTI-HAL: A Human-Annotated Dataset for Cyber Threat Intelligence Analysis ABSTRACT: Organizations are increasingly targeted by Advanced Persistent Threats (APTs), which involve complex, multi-stage tactics and diverse techniques. Cyber Threat Intelligence (CTI) sources, such as incident reports and securit...
2504.05878
Xingyuan Li
Xingyuan Li, Ruichao Hou, Tongwei Ren, Gangshan Wu
KAN-SAM: Kolmogorov-Arnold Network Guided Segment Anything Model for RGB-T Salient Object Detection
This paper is accepted by ICME2025
null
null
null
cs.MM cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Existing RGB-thermal salient object detection (RGB-T SOD) methods aim to identify visually significant objects by leveraging both RGB and thermal modalities to enable robust performance in complex scenarios, but they often suffer from limited generalization due to the constrained diversity of available datasets and t...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 10:07:02 GMT" } ]
2025-04-09T00:00:00
[ [ "Li", "Xingyuan", "" ], [ "Hou", "Ruichao", "" ], [ "Ren", "Tongwei", "" ], [ "Wu", "Gangshan", "" ] ]
TITLE: KAN-SAM: Kolmogorov-Arnold Network Guided Segment Anything Model for RGB-T Salient Object Detection ABSTRACT: Existing RGB-thermal salient object detection (RGB-T SOD) methods aim to identify visually significant objects by leveraging both RGB and thermal modalities to enable robust performance in complex ...
2504.05882
Luca Barco
Luca Barco, Giacomo Blanco, Gaetano Chiriaco, Alessia Intini, Luigi La Riccia, Vittorio Scolamiero, Piero Boccardo, Paolo Garza, Fabrizio Dominici
Turin3D: Evaluating Adaptation Strategies under Label Scarcity in Urban LiDAR Segmentation with Semi-Supervised Techniques
Accepted at CVPRW2025 - USM3D
null
null
null
cs.CV cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
3D semantic segmentation plays a critical role in urban modelling, enabling detailed understanding and mapping of city environments. In this paper, we introduce Turin3D: a new aerial LiDAR dataset for point cloud semantic segmentation covering an area of around 1.43 km2 in the city centre of Turin with almost 70M poi...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 10:17:14 GMT" } ]
2025-04-09T00:00:00
[ [ "Barco", "Luca", "" ], [ "Blanco", "Giacomo", "" ], [ "Chiriaco", "Gaetano", "" ], [ "Intini", "Alessia", "" ], [ "La Riccia", "Luigi", "" ], [ "Scolamiero", "Vittorio", "" ], [ "Boccardo", "Piero", "" ],...
TITLE: Turin3D: Evaluating Adaptation Strategies under Label Scarcity in Urban LiDAR Segmentation with Semi-Supervised Techniques ABSTRACT: 3D semantic segmentation plays a critical role in urban modelling, enabling detailed understanding and mapping of city environments. In this paper, we introduce Turin3D: a ne...
2504.05888
Guillaume Gautier
Guillaume Gautier, Alexandre Mercat, Louis Fr\'eneau, Mikko Pitk\"anen, and Jarno Vanne
UVG-VPC: Voxelized Point Cloud Dataset for Visual Volumetric Video-based Coding
Point cloud compression;Geometry;Visualization;Three-dimensional displays;Video sequences;Transform coding;Media;Open dataset;point cloud;Visual Volumetric Video-based Coding (V3C);Video-based Point Cloud Compression (V-PCC);Extended Reality (XR)
2023 15th International Conference on Quality of Multimedia Experience (QoMEX), Ghent, Belgium, 2023, pp. 244-247
10.1109/QoMEX58391.2023.10178589
null
cs.MM cs.CV
http://creativecommons.org/licenses/by-nc-nd/4.0/
Point cloud compression has become a crucial factor in immersive visual media processing and streaming. This paper presents a new open dataset called UVG-VPC for the development, evaluation, and validation of MPEG Visual Volumetric Video-based Coding (V3C) technology. The dataset is distributed under its own non-comm...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 10:27:53 GMT" } ]
2025-04-09T00:00:00
[ [ "Gautier", "Guillaume", "" ], [ "Mercat", "Alexandre", "" ], [ "Fréneau", "Louis", "" ], [ "Pitkänen", "Mikko", "" ], [ "Vanne", "Jarno", "" ] ]
TITLE: UVG-VPC: Voxelized Point Cloud Dataset for Visual Volumetric Video-based Coding ABSTRACT: Point cloud compression has become a crucial factor in immersive visual media processing and streaming. This paper presents a new open dataset called UVG-VPC for the development, evaluation, and validation of MPEG Vis...
2504.05894
Ivan Svetunkov
Ivan Svetunkov and Anna Sroginis
Why do zeroes happen? A model-based approach for demand classification
39 pages, 11 figures, 3 tables
null
null
null
cs.LG stat.ME
http://creativecommons.org/licenses/by-nc-sa/4.0/
Effective demand forecasting is critical for inventory management, production planning, and decision making across industries. Selecting the appropriate model and suitable features to efficiently capture patterns in the data is one of the main challenges in demand forecasting. In reality, this becomes even more compl...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 10:45:30 GMT" } ]
2025-04-09T00:00:00
[ [ "Svetunkov", "Ivan", "" ], [ "Sroginis", "Anna", "" ] ]
TITLE: Why do zeroes happen? A model-based approach for demand classification ABSTRACT: Effective demand forecasting is critical for inventory management, production planning, and decision making across industries. Selecting the appropriate model and suitable features to efficiently capture patterns in the data is ...
2504.05904
Xiangyu Zheng
Xiangyu Zheng, Wanyun Li, Songcheng He, Xiaoqiang Li, We Zhang
Intrinsic Saliency Guided Trunk-Collateral Network for Unsupervised Video Object Segmentation
null
null
null
null
cs.CV cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Recent unsupervised video object segmentation (UVOS) methods predominantly adopt the motion-appearance paradigm. Mainstream motion-appearance approaches use either the two-encoder structure to separately encode motion and appearance features, or the single-encoder structure for joint encoding. However, these methods ...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 11:02:14 GMT" } ]
2025-04-09T00:00:00
[ [ "Zheng", "Xiangyu", "" ], [ "Li", "Wanyun", "" ], [ "He", "Songcheng", "" ], [ "Li", "Xiaoqiang", "" ], [ "Zhang", "We", "" ] ]
TITLE: Intrinsic Saliency Guided Trunk-Collateral Network for Unsupervised Video Object Segmentation ABSTRACT: Recent unsupervised video object segmentation (UVOS) methods predominantly adopt the motion-appearance paradigm. Mainstream motion-appearance approaches use either the two-encoder structure to separately...
2504.05908
Sriram Mandalika
Sriram Mandalika, Lalitha V, Athira Nambiar
PRIMEDrive-CoT: A Precognitive Chain-of-Thought Framework for Uncertainty-Aware Object Interaction in Driving Scene Scenario
Accepted at The IEEE/CVF Conference on Computer Vision and Pattern Recognition 2025 - CVPRW
null
null
null
cs.CV cs.AI cs.LG
http://creativecommons.org/licenses/by/4.0/
Driving scene understanding is a critical real-world problem that involves interpreting and associating various elements of a driving environment, such as vehicles, pedestrians, and traffic signals. Despite advancements in autonomous driving, traditional pipelines rely on deterministic models that fail to capture the...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 11:06:02 GMT" } ]
2025-04-09T00:00:00
[ [ "Mandalika", "Sriram", "" ], [ "V", "Lalitha", "" ], [ "Nambiar", "Athira", "" ] ]
TITLE: PRIMEDrive-CoT: A Precognitive Chain-of-Thought Framework for Uncertainty-Aware Object Interaction in Driving Scene Scenario ABSTRACT: Driving scene understanding is a critical real-world problem that involves interpreting and associating various elements of a driving environment, such as vehicles, pedestr...
2504.05914
Abhiram Reddy Yanampally
Abhiram Reddy Yanampally
High-Resource Translation:Turning Abundance into Accessibility
6 pages, 2 figures
null
null
null
cs.CL
http://creativecommons.org/licenses/by/4.0/
This paper presents a novel approach to constructing an English-to-Telugu translation model by leveraging transfer learning techniques and addressing the challenges associated with low-resource languages. Utilizing the Bharat Parallel Corpus Collection (BPCC) as the primary dataset, the model incorporates iterative b...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 11:09:51 GMT" } ]
2025-04-09T00:00:00
[ [ "Yanampally", "Abhiram Reddy", "" ] ]
TITLE: High-Resource Translation:Turning Abundance into Accessibility ABSTRACT: This paper presents a novel approach to constructing an English-to-Telugu translation model by leveraging transfer learning techniques and addressing the challenges associated with low-resource languages. Utilizing the Bharat Parallel C...
2504.05917
Solon Pissis
Giulia Bernardini and Huiping Chen and Alessio Conte and Roberto Grossi and Veronica Guerrini and Grigorios Loukides and Nadia Pisanti and and Solon P. Pissis
Indexing Strings with Utilities
ICDE 2025 (abstract abridged to satisfy arXiv requirements)
null
null
null
cs.DS cs.DB
http://creativecommons.org/licenses/by/4.0/
Applications in domains ranging from bioinformatics to advertising feature strings that come with numerical scores (utilities). The utilities quantify the importance, interest, profit, or risk of the letters occurring at every position of a string. Motivated by the ever-increasing rate of generating such data, as wel...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 11:13:53 GMT" } ]
2025-04-09T00:00:00
[ [ "Bernardini", "Giulia", "" ], [ "Chen", "Huiping", "" ], [ "Conte", "Alessio", "" ], [ "Grossi", "Roberto", "" ], [ "Guerrini", "Veronica", "" ], [ "Loukides", "Grigorios", "" ], [ "Pisanti", "Nadia", "" ...
TITLE: Indexing Strings with Utilities ABSTRACT: Applications in domains ranging from bioinformatics to advertising feature strings that come with numerical scores (utilities). The utilities quantify the importance, interest, profit, or risk of the letters occurring at every position of a string. Motivated by the e...
2504.05923
Juliett Su\'arez Ferreira
Juliett Su\'arez Ferreira, Marija Slavkovik, Jorge Casillas
Uncovering Fairness through Data Complexity as an Early Indicator
null
null
null
null
cs.LG cs.AI cs.DS
http://creativecommons.org/licenses/by-nc-nd/4.0/
Fairness constitutes a concern within machine learning (ML) applications. Currently, there is no study on how disparities in classification complexity between privileged and unprivileged groups could influence the fairness of solutions, which serves as a preliminary indicator of potential unfairness. In this work, we...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 11:28:40 GMT" } ]
2025-04-09T00:00:00
[ [ "Ferreira", "Juliett Suárez", "" ], [ "Slavkovik", "Marija", "" ], [ "Casillas", "Jorge", "" ] ]
TITLE: Uncovering Fairness through Data Complexity as an Early Indicator ABSTRACT: Fairness constitutes a concern within machine learning (ML) applications. Currently, there is no study on how disparities in classification complexity between privileged and unprivileged groups could influence the fairness of solutio...
2504.05957
Julian Agudelo
Julian Agudelo and Vincent Guigue and Cristina Manfredotti and Hadrien Piot
Drought forecasting using a hybrid neural architecture for integrating time series and static data
10 pages, 3 figures, published as a workshop paper at Tackling Climate Change with Machine Learning at ICLR 2025, Tackling Climate Change with Machine Learning is a non-archival workshop
null
null
null
cs.LG
http://creativecommons.org/licenses/by-nc-sa/4.0/
Reliable forecasting is critical for early warning systems and adaptive drought management. Most previous deep learning approaches focus solely on homogeneous regions and rely on single-structured data. This paper presents a hybrid neural architecture that integrates time series and static data, achieving state-of-th...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 12:11:34 GMT" } ]
2025-04-09T00:00:00
[ [ "Agudelo", "Julian", "" ], [ "Guigue", "Vincent", "" ], [ "Manfredotti", "Cristina", "" ], [ "Piot", "Hadrien", "" ] ]
TITLE: Drought forecasting using a hybrid neural architecture for integrating time series and static data ABSTRACT: Reliable forecasting is critical for early warning systems and adaptive drought management. Most previous deep learning approaches focus solely on homogeneous regions and rely on single-structured d...
2504.05966
Xiaolin Fan
Xiaolin Fan, Yan Wang, Yingying Zhang, Mingkun Bao, Bosen Jia, Dong Lu, Yifan Gu, Jian Cheng, and Haogang Zhu
AVP-AP: Self-supervised Automatic View Positioning in 3D cardiac CT via Atlas Prompting
12 pages, 8 figures, published to TMI
IEEE TRANSACTIONS ON MEDICAL IMAGING, March 2025
10.1109/TMI.2025.3554785
null
eess.IV cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Automatic view positioning is crucial for cardiac computed tomography (CT) examinations, including disease diagnosis and surgical planning. However, it is highly challenging due to individual variability and large 3D search space. Existing work needs labor-intensive and time-consuming manual annotations to train view...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 12:24:37 GMT" } ]
2025-04-09T00:00:00
[ [ "Fan", "Xiaolin", "" ], [ "Wang", "Yan", "" ], [ "Zhang", "Yingying", "" ], [ "Bao", "Mingkun", "" ], [ "Jia", "Bosen", "" ], [ "Lu", "Dong", "" ], [ "Gu", "Yifan", "" ], [ "Cheng", "Jian", ...
TITLE: AVP-AP: Self-supervised Automatic View Positioning in 3D cardiac CT via Atlas Prompting ABSTRACT: Automatic view positioning is crucial for cardiac computed tomography (CT) examinations, including disease diagnosis and surgical planning. However, it is highly challenging due to individual variability and l...
2504.05977
Jakob Christensen
Jakob L{\o}nborg Christensen, Morten Rieger Hannemose, Anders Bjorholm Dahl, Vedrana Andersen Dahl
Diffusion Based Ambiguous Image Segmentation
Accepted at SCIA25
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Medical image segmentation often involves inherent uncertainty due to variations in expert annotations. Capturing this uncertainty is an important goal and previous works have used various generative image models for the purpose of representing the full distribution of plausible expert ground truths. In this work, we...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 12:33:26 GMT" } ]
2025-04-09T00:00:00
[ [ "Christensen", "Jakob Lønborg", "" ], [ "Hannemose", "Morten Rieger", "" ], [ "Dahl", "Anders Bjorholm", "" ], [ "Dahl", "Vedrana Andersen", "" ] ]
TITLE: Diffusion Based Ambiguous Image Segmentation ABSTRACT: Medical image segmentation often involves inherent uncertainty due to variations in expert annotations. Capturing this uncertainty is an important goal and previous works have used various generative image models for the purpose of representing the full ...
2504.05992
Jie Yang
Jie Yang, Chang Su, Yuhan Zhang, Jianjun Zhu and Jianli Wang
Under-Sampled High-Dimensional Data Recovery via Symbiotic Multi-Prior Tensor Reconstruction
null
null
null
null
eess.IV cs.CV
http://creativecommons.org/licenses/by/4.0/
The advancement of sensing technology has driven the widespread application of high-dimensional data. However, issues such as missing entries during acquisition and transmission negatively impact the accuracy of subsequent tasks. Tensor reconstruction aims to recover the underlying complete data from under-sampled ob...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 12:55:18 GMT" } ]
2025-04-09T00:00:00
[ [ "Yang", "Jie", "" ], [ "Su", "Chang", "" ], [ "Zhang", "Yuhan", "" ], [ "Zhu", "Jianjun", "" ], [ "Wang", "Jianli", "" ] ]
TITLE: Under-Sampled High-Dimensional Data Recovery via Symbiotic Multi-Prior Tensor Reconstruction ABSTRACT: The advancement of sensing technology has driven the widespread application of high-dimensional data. However, issues such as missing entries during acquisition and transmission negatively impact the accu...
2504.05995
Firoj Alam
Firoj Alam, Md Arid Hasan, Sahinur Rahman Laskar, Mucahid Kutlu, Shammur Absar Chowdhury
NativQA Framework: Enabling LLMs with Native, Local, and Everyday Knowledge
LLMs, Native, Multilingual, Language Diversity, Contextual Understanding, Minority Languages, Culturally Informed, Foundation Models, Large Language Models
null
null
null
cs.CL cs.AI
http://creativecommons.org/licenses/by-nc-sa/4.0/
The rapid advancement of large language models (LLMs) has raised concerns about cultural bias, fairness, and their applicability in diverse linguistic and underrepresented regional contexts. To enhance and benchmark the capabilities of LLMs, there is a need to develop large-scale resources focused on multilingual, lo...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 13:01:51 GMT" } ]
2025-04-09T00:00:00
[ [ "Alam", "Firoj", "" ], [ "Hasan", "Md Arid", "" ], [ "Laskar", "Sahinur Rahman", "" ], [ "Kutlu", "Mucahid", "" ], [ "Chowdhury", "Shammur Absar", "" ] ]
TITLE: NativQA Framework: Enabling LLMs with Native, Local, and Everyday Knowledge ABSTRACT: The rapid advancement of large language models (LLMs) has raised concerns about cultural bias, fairness, and their applicability in diverse linguistic and underrepresented regional contexts. To enhance and benchmark the c...
2504.06003
Can Zhang
Can Zhang and Gim Hee Lee
econSG: Efficient and Multi-view Consistent Open-Vocabulary 3D Semantic Gaussians
null
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The primary focus of most recent works on open-vocabulary neural fields is extracting precise semantic features from the VLMs and then consolidating them efficiently into a multi-view consistent 3D neural fields representation. However, most existing works over-trusted SAM to regularize image-level CLIP without any f...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 13:12:31 GMT" } ]
2025-04-09T00:00:00
[ [ "Zhang", "Can", "" ], [ "Lee", "Gim Hee", "" ] ]
TITLE: econSG: Efficient and Multi-view Consistent Open-Vocabulary 3D Semantic Gaussians ABSTRACT: The primary focus of most recent works on open-vocabulary neural fields is extracting precise semantic features from the VLMs and then consolidating them efficiently into a multi-view consistent 3D neural fields rep...
2504.06004
Mrityunjoy Gain
Mrityunjoy Gain, Kitae Kim, Avi Deb Raha, Apurba Adhikary, Eui-Nam Huh, Zhu Han, and Choong Seon Hong
FedFeat+: A Robust Federated Learning Framework Through Federated Aggregation and Differentially Private Feature-Based Classifier Retraining
null
null
null
null
cs.CV
http://creativecommons.org/licenses/by-nc-nd/4.0/
In this paper, we propose the FedFeat+ framework, which distinctively separates feature extraction from classification. We develop a two-tiered model training process: following local training, clients transmit their weights and some features extracted from the feature extractor from the final local epochs to the ser...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 13:12:38 GMT" } ]
2025-04-09T00:00:00
[ [ "Gain", "Mrityunjoy", "" ], [ "Kim", "Kitae", "" ], [ "Raha", "Avi Deb", "" ], [ "Adhikary", "Apurba", "" ], [ "Huh", "Eui-Nam", "" ], [ "Han", "Zhu", "" ], [ "Hong", "Choong Seon", "" ] ]
TITLE: FedFeat+: A Robust Federated Learning Framework Through Federated Aggregation and Differentially Private Feature-Based Classifier Retraining ABSTRACT: In this paper, we propose the FedFeat+ framework, which distinctively separates feature extraction from classification. We develop a two-tiered model traini...
2504.06006
Roman Kochnev
Roman Kochnev, Arash Torabi Goodarzi, Zofia Antonina Bentyn, Dmitry Ignatov, Radu Timofte
Optuna vs Code Llama: Are LLMs a New Paradigm for Hyperparameter Tuning?
null
null
null
null
cs.LG cs.AI cs.NE
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Optimal hyperparameter selection is critical for maximizing neural network performance, especially as models grow in complexity. This work investigates the viability of using large language models (LLMs) for hyperparameter optimization by employing a fine-tuned version of Code Llama. Through parameter-efficient fine-...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 13:15:47 GMT" } ]
2025-04-09T00:00:00
[ [ "Kochnev", "Roman", "" ], [ "Goodarzi", "Arash Torabi", "" ], [ "Bentyn", "Zofia Antonina", "" ], [ "Ignatov", "Dmitry", "" ], [ "Timofte", "Radu", "" ] ]
TITLE: Optuna vs Code Llama: Are LLMs a New Paradigm for Hyperparameter Tuning? ABSTRACT: Optimal hyperparameter selection is critical for maximizing neural network performance, especially as models grow in complexity. This work investigates the viability of using large language models (LLMs) for hyperparameter opt...
2504.06010
Stefanos-Iordanis Papadopoulos
Stefanos-Iordanis Papadopoulos, Christos Koutlis, Symeon Papadopoulos, Panagiotis C. Petrantonakis
Latent Multimodal Reconstruction for Misinformation Detection
null
null
null
null
cs.CV cs.MM
http://creativecommons.org/licenses/by-sa/4.0/
Multimodal misinformation, such as miscaptioned images, where captions misrepresent an image's origin, context, or meaning, poses a growing challenge in the digital age. To support fact-checkers, researchers have been focusing on creating datasets and developing methods for multimodal misinformation detection (MMD). ...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 13:16:48 GMT" } ]
2025-04-09T00:00:00
[ [ "Papadopoulos", "Stefanos-Iordanis", "" ], [ "Koutlis", "Christos", "" ], [ "Papadopoulos", "Symeon", "" ], [ "Petrantonakis", "Panagiotis C.", "" ] ]
TITLE: Latent Multimodal Reconstruction for Misinformation Detection ABSTRACT: Multimodal misinformation, such as miscaptioned images, where captions misrepresent an image's origin, context, or meaning, poses a growing challenge in the digital age. To support fact-checkers, researchers have been focusing on creatin...
2504.06022
Luis Denninger
Luis Denninger, Sina Mokhtarzadeh Azar, Juergen Gall
CamContextI2V: Context-aware Controllable Video Generation
null
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Recently, image-to-video (I2V) diffusion models have demonstrated impressive scene understanding and generative quality, incorporating image conditions to guide generation. However, these models primarily animate static images without extending beyond their provided context. Introducing additional constraints, such a...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 13:26:59 GMT" } ]
2025-04-09T00:00:00
[ [ "Denninger", "Luis", "" ], [ "Azar", "Sina Mokhtarzadeh", "" ], [ "Gall", "Juergen", "" ] ]
TITLE: CamContextI2V: Context-aware Controllable Video Generation ABSTRACT: Recently, image-to-video (I2V) diffusion models have demonstrated impressive scene understanding and generative quality, incorporating image conditions to guide generation. However, these models primarily animate static images without exten...
2504.06039
Julia Werner
Julia Werner, Christoph Gerum, Jorg Nick, Maxime Le Floch, Franz Brinkmann, Jochen Hampe, and Oliver Bringmann
Enhanced Anomaly Detection for Capsule Endoscopy Using Ensemble Learning Strategies
Accepted at the 47th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBS EMBC)
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Capsule endoscopy is a method to capture images of the gastrointestinal tract and screen for diseases which might remain hidden if investigated with standard endoscopes. Due to the limited size of a video capsule, embedding AI models directly into the capsule demands careful consideration of the model size and thus c...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 13:39:39 GMT" } ]
2025-04-09T00:00:00
[ [ "Werner", "Julia", "" ], [ "Gerum", "Christoph", "" ], [ "Nick", "Jorg", "" ], [ "Floch", "Maxime Le", "" ], [ "Brinkmann", "Franz", "" ], [ "Hampe", "Jochen", "" ], [ "Bringmann", "Oliver", "" ] ]
TITLE: Enhanced Anomaly Detection for Capsule Endoscopy Using Ensemble Learning Strategies ABSTRACT: Capsule endoscopy is a method to capture images of the gastrointestinal tract and screen for diseases which might remain hidden if investigated with standard endoscopes. Due to the limited size of a video capsule,...
2504.06055
Panagiota Rempi
Panagiota Rempi, Sotiris Pelekis, Alexandros Menelaos Tzortzis, Evangelos Karakolis, Christos Ntanos, Dimitris Askounis
Explainable AI for building energy retrofitting under data scarcity
null
null
null
null
cs.LG
http://creativecommons.org/licenses/by/4.0/
Enhancing energy efficiency in residential buildings is a crucial step toward mitigating climate change and reducing greenhouse gas emissions. Retrofitting existing buildings, which account for a significant portion of energy consumption, is critical particularly in regions with outdated and inefficient building stoc...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 14:00:08 GMT" } ]
2025-04-09T00:00:00
[ [ "Rempi", "Panagiota", "" ], [ "Pelekis", "Sotiris", "" ], [ "Tzortzis", "Alexandros Menelaos", "" ], [ "Karakolis", "Evangelos", "" ], [ "Ntanos", "Christos", "" ], [ "Askounis", "Dimitris", "" ] ]
TITLE: Explainable AI for building energy retrofitting under data scarcity ABSTRACT: Enhancing energy efficiency in residential buildings is a crucial step toward mitigating climate change and reducing greenhouse gas emissions. Retrofitting existing buildings, which account for a significant portion of energy consu...
2504.06069
Reza Masoudian Saadabad
Hanieh Masoudian Saadabad, Lingraj Kumar, Reza Masoudian Saadabad, and Maja Colautti
Physics-Constrained Neural Network for Metasurface Optical Response Prediction
null
null
null
null
physics.optics
http://creativecommons.org/licenses/by/4.0/
A physics-constrained neural network is presented for predicting the optical response of metasurfaces. Our approach incorporates physical laws directly into the neural network architecture and loss function, addressing critical challenges in the modeling of metasurfaces. Unlike methods that require specialized weight...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 14:10:28 GMT" } ]
2025-04-09T00:00:00
[ [ "Saadabad", "Hanieh Masoudian", "" ], [ "Kumar", "Lingraj", "" ], [ "Saadabad", "Reza Masoudian", "" ], [ "Colautti", "Maja", "" ] ]
TITLE: Physics-Constrained Neural Network for Metasurface Optical Response Prediction ABSTRACT: A physics-constrained neural network is presented for predicting the optical response of metasurfaces. Our approach incorporates physical laws directly into the neural network architecture and loss function, addressing...
2504.06084
Alexey Gavryushin
Alexey Gavryushin, Xi Wang, Robert J. S. Malate, Chenyu Yang, Xiangyi Jia, Shubh Goel, Davide Liconti, Ren\'e Zurbr\"ugg, Robert K. Katzschmann, Marc Pollefeys
MAPLE: Encoding Dexterous Robotic Manipulation Priors Learned From Egocentric Videos
null
null
null
null
cs.RO cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Large-scale egocentric video datasets capture diverse human activities across a wide range of scenarios, offering rich and detailed insights into how humans interact with objects, especially those that require fine-grained dexterous control. Such complex, dexterous skills with precise controls are crucial for many ro...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 14:25:25 GMT" } ]
2025-04-09T00:00:00
[ [ "Gavryushin", "Alexey", "" ], [ "Wang", "Xi", "" ], [ "Malate", "Robert J. S.", "" ], [ "Yang", "Chenyu", "" ], [ "Jia", "Xiangyi", "" ], [ "Goel", "Shubh", "" ], [ "Liconti", "Davide", "" ], [ "Zur...
TITLE: MAPLE: Encoding Dexterous Robotic Manipulation Priors Learned From Egocentric Videos ABSTRACT: Large-scale egocentric video datasets capture diverse human activities across a wide range of scenarios, offering rich and detailed insights into how humans interact with objects, especially those that require fi...
2504.06088
Pramit Saha
Divyanshu Mishra, Pramit Saha, He Zhao, Netzahualcoyotl Hernandez-Cruz, Olga Patey, Aris Papageorghiou, J. Alison Noble
MCAT: Visual Query-Based Localization of Standard Anatomical Clips in Fetal Ultrasound Videos Using Multi-Tier Class-Aware Token Transformer
Accepted in AAAI 2025
null
null
null
cs.CV cs.AI cs.LG
http://creativecommons.org/licenses/by/4.0/
Accurate standard plane acquisition in fetal ultrasound (US) videos is crucial for fetal growth assessment, anomaly detection, and adherence to clinical guidelines. However, manually selecting standard frames is time-consuming and prone to intra- and inter-sonographer variability. Existing methods primarily rely on i...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 14:29:15 GMT" } ]
2025-04-09T00:00:00
[ [ "Mishra", "Divyanshu", "" ], [ "Saha", "Pramit", "" ], [ "Zhao", "He", "" ], [ "Hernandez-Cruz", "Netzahualcoyotl", "" ], [ "Patey", "Olga", "" ], [ "Papageorghiou", "Aris", "" ], [ "Noble", "J. Alison", ""...
TITLE: MCAT: Visual Query-Based Localization of Standard Anatomical Clips in Fetal Ultrasound Videos Using Multi-Tier Class-Aware Token Transformer ABSTRACT: Accurate standard plane acquisition in fetal ultrasound (US) videos is crucial for fetal growth assessment, anomaly detection, and adherence to clinical gui...
2504.06099
\v{S}imon Bil\'ik
Samuel Bielik, Simon Bilik
Towards Varroa destructor mite detection using a narrow spectra illumination
null
null
null
null
cs.CV cs.AI cs.LG
http://creativecommons.org/licenses/by/4.0/
This paper focuses on the development and modification of a beehive monitoring device and Varroa destructor detection on the bees with the help of hyperspectral imagery while utilizing a U-net, semantic segmentation architecture, and conventional computer vision methods. The main objectives were to collect a dataset ...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 14:41:42 GMT" } ]
2025-04-09T00:00:00
[ [ "Bielik", "Samuel", "" ], [ "Bilik", "Simon", "" ] ]
TITLE: Towards Varroa destructor mite detection using a narrow spectra illumination ABSTRACT: This paper focuses on the development and modification of a beehive monitoring device and Varroa destructor detection on the bees with the help of hyperspectral imagery while utilizing a U-net, semantic segmentation arch...
2504.06102
Eric Wagner
Eric Wagner and Lennart Bader and Konrad Wolsing and Martin Serror
Sherlock: A Dataset for Process-aware Intrusion Detection Research on Power Grid Networks
accepted at CODASPY'25
null
10.1145/3714393.3726006
null
cs.CR cs.NI
http://creativecommons.org/licenses/by/4.0/
Physically distributed components and legacy protocols make the protection of power grids against increasing cyberattack threats challenging. Infamously, the 2015 and 2016 blackouts in Ukraine were caused by cyberattacks, and the German Federal Office for Information Security (BSI) recorded over 200 cyber incidents a...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 14:46:35 GMT" } ]
2025-04-09T00:00:00
[ [ "Wagner", "Eric", "" ], [ "Bader", "Lennart", "" ], [ "Wolsing", "Konrad", "" ], [ "Serror", "Martin", "" ] ]
TITLE: Sherlock: A Dataset for Process-aware Intrusion Detection Research on Power Grid Networks ABSTRACT: Physically distributed components and legacy protocols make the protection of power grids against increasing cyberattack threats challenging. Infamously, the 2015 and 2016 blackouts in Ukraine were caused by...
2504.06105
Abinav Kalyanasundaram
Abinav Kalyanasundaram, Karthikeyan Chandra Sekaran, Philipp Stauber, Michael Lange, Wolfgang Utschick and Michael Botsch
Uncertainty-Aware Hybrid Machine Learning in Virtual Sensors for Vehicle Sideslip Angle Estimation
Accepted at the 2025 IEEE Intelligent Vehicles Symposium (IV)
null
null
null
cs.RO cs.AI cs.LG
http://creativecommons.org/licenses/by/4.0/
Precise vehicle state estimation is crucial for safe and reliable autonomous driving. The number of measurable states and their precision offered by the onboard vehicle sensor system are often constrained by cost. For instance, measuring critical quantities such as the Vehicle Sideslip Angle (VSA) poses significant c...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 14:49:58 GMT" } ]
2025-04-09T00:00:00
[ [ "Kalyanasundaram", "Abinav", "" ], [ "Sekaran", "Karthikeyan Chandra", "" ], [ "Stauber", "Philipp", "" ], [ "Lange", "Michael", "" ], [ "Utschick", "Wolfgang", "" ], [ "Botsch", "Michael", "" ] ]
TITLE: Uncertainty-Aware Hybrid Machine Learning in Virtual Sensors for Vehicle Sideslip Angle Estimation ABSTRACT: Precise vehicle state estimation is crucial for safe and reliable autonomous driving. The number of measurable states and their precision offered by the onboard vehicle sensor system are often const...
2504.06116
Davide Sferrazza
Davide Sferrazza, Gabriele Berton, Gabriele Trivigno, Carlo Masone
To Match or Not to Match: Revisiting Image Matching for Reliable Visual Place Recognition
CVPRW 2025
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Visual Place Recognition (VPR) is a critical task in computer vision, traditionally enhanced by re-ranking retrieval results with image matching. However, recent advancements in VPR methods have significantly improved performance, challenging the necessity of re-ranking. In this work, we show that modern retrieval sy...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 15:10:10 GMT" } ]
2025-04-09T00:00:00
[ [ "Sferrazza", "Davide", "" ], [ "Berton", "Gabriele", "" ], [ "Trivigno", "Gabriele", "" ], [ "Masone", "Carlo", "" ] ]
TITLE: To Match or Not to Match: Revisiting Image Matching for Reliable Visual Place Recognition ABSTRACT: Visual Place Recognition (VPR) is a critical task in computer vision, traditionally enhanced by re-ranking retrieval results with image matching. However, recent advancements in VPR methods have significantl...
2504.06120
Yuanpei Liu
Yuanpei Liu, Zhenqi He, Kai Han
Hyperbolic Category Discovery
Accepted as a conference paper at CVPR 2025
null
null
null
cs.CV
http://creativecommons.org/licenses/by-nc-sa/4.0/
Generalized Category Discovery (GCD) is an intriguing open-world problem that has garnered increasing attention. Given a dataset that includes both labelled and unlabelled images, GCD aims to categorize all images in the unlabelled subset, regardless of whether they belong to known or unknown classes. In GCD, the com...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 15:12:33 GMT" } ]
2025-04-09T00:00:00
[ [ "Liu", "Yuanpei", "" ], [ "He", "Zhenqi", "" ], [ "Han", "Kai", "" ] ]
TITLE: Hyperbolic Category Discovery ABSTRACT: Generalized Category Discovery (GCD) is an intriguing open-world problem that has garnered increasing attention. Given a dataset that includes both labelled and unlabelled images, GCD aims to categorize all images in the unlabelled subset, regardless of whether they be...
2504.06121
Yuhang Ma
Ronghui Zhang, Yuhang Ma, Tengfei Li, Ziyu Lin, Yueying Wu, Junzhou Chen, Lin Zhang, Jia Hu, Tony Z. Qiu and Konghui Guo
A Robust Real-Time Lane Detection Method with Fog-Enhanced Feature Fusion for Foggy Conditions
null
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Lane detection is a critical component of Advanced Driver Assistance Systems (ADAS). Existing lane detection algorithms generally perform well under favorable weather conditions. However, their performance degrades significantly in adverse conditions, such as fog, which increases the risk of traffic accidents. This c...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 15:13:01 GMT" } ]
2025-04-09T00:00:00
[ [ "Zhang", "Ronghui", "" ], [ "Ma", "Yuhang", "" ], [ "Li", "Tengfei", "" ], [ "Lin", "Ziyu", "" ], [ "Wu", "Yueying", "" ], [ "Chen", "Junzhou", "" ], [ "Zhang", "Lin", "" ], [ "Hu", "Jia", "...
TITLE: A Robust Real-Time Lane Detection Method with Fog-Enhanced Feature Fusion for Foggy Conditions ABSTRACT: Lane detection is a critical component of Advanced Driver Assistance Systems (ADAS). Existing lane detection algorithms generally perform well under favorable weather conditions. However, their performa...
2504.06136
Movina Moses
Movina Moses, Mohab Elkaref, James Barry, Shinnosuke Tanaka, Vishnudev Kuruvanthodi, Nathan Herr, Campbell D Watson, Geeth De Mel
QGen Studio: An Adaptive Question-Answer Generation, Training and Evaluation Platform
null
null
null
null
cs.CL cs.AI
http://creativecommons.org/licenses/by-nc-nd/4.0/
We present QGen Studio: an adaptive question-answer generation, training, and evaluation platform. QGen Studio enables users to leverage large language models (LLMs) to create custom question-answer datasets and fine-tune models on this synthetic data. It features a dataset viewer and model explorer to streamline thi...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 15:32:09 GMT" } ]
2025-04-09T00:00:00
[ [ "Moses", "Movina", "" ], [ "Elkaref", "Mohab", "" ], [ "Barry", "James", "" ], [ "Tanaka", "Shinnosuke", "" ], [ "Kuruvanthodi", "Vishnudev", "" ], [ "Herr", "Nathan", "" ], [ "Watson", "Campbell D", "" ]...
TITLE: QGen Studio: An Adaptive Question-Answer Generation, Training and Evaluation Platform ABSTRACT: We present QGen Studio: an adaptive question-answer generation, training, and evaluation platform. QGen Studio enables users to leverage large language models (LLMs) to create custom question-answer datasets and...
2504.06148
Xiangxi Zheng
Xiangxi Zheng, Linjie Li, Zhengyuan Yang, Ping Yu, Alex Jinpeng Wang, Rui Yan, Yuan Yao, Lijuan Wang
V-MAGE: A Game Evaluation Framework for Assessing Visual-Centric Capabilities in Multimodal Large Language Models
null
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Recent advancements in Multimodal Large Language Models (MLLMs) have led to significant improvements across various multimodal benchmarks. However, as evaluations shift from static datasets to open-world, dynamic environments, current game-based benchmarks remain inadequate because they lack visual-centric tasks and ...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 15:43:01 GMT" } ]
2025-04-09T00:00:00
[ [ "Zheng", "Xiangxi", "" ], [ "Li", "Linjie", "" ], [ "Yang", "Zhengyuan", "" ], [ "Yu", "Ping", "" ], [ "Wang", "Alex Jinpeng", "" ], [ "Yan", "Rui", "" ], [ "Yao", "Yuan", "" ], [ "Wang", "Lijua...
TITLE: V-MAGE: A Game Evaluation Framework for Assessing Visual-Centric Capabilities in Multimodal Large Language Models ABSTRACT: Recent advancements in Multimodal Large Language Models (MLLMs) have led to significant improvements across various multimodal benchmarks. However, as evaluations shift from static da...
2504.06153
Akash Kumar
Akash Kumar, Ashlesha Kumar, Vibhav Vineet, Yogesh S Rawat
A Large-Scale Analysis on Contextual Self-Supervised Video Representation Learning
CVPR'25 Workshop: 6th Data-Efficient Workshop
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Self-supervised learning has emerged as a powerful paradigm for label-free model pretraining, particularly in the video domain, where manual annotation is costly and time-intensive. However, existing self-supervised approaches employ diverse experimental setups, making direct comparisons challenging due to the absenc...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 15:47:58 GMT" } ]
2025-04-09T00:00:00
[ [ "Kumar", "Akash", "" ], [ "Kumar", "Ashlesha", "" ], [ "Vineet", "Vibhav", "" ], [ "Rawat", "Yogesh S", "" ] ]
TITLE: A Large-Scale Analysis on Contextual Self-Supervised Video Representation Learning ABSTRACT: Self-supervised learning has emerged as a powerful paradigm for label-free model pretraining, particularly in the video domain, where manual annotation is costly and time-intensive. However, existing self-supervise...
2504.06156
Chuanyu Li
Fangchen Liu, Chuanyu Li, Yihua Qin, Ankit Shaw, Jing Xu, Pieter Abbeel, Rui Chen
ViTaMIn: Learning Contact-Rich Tasks Through Robot-Free Visuo-Tactile Manipulation Interface
null
null
null
null
cs.RO
http://creativecommons.org/licenses/by/4.0/
Tactile information plays a crucial role for humans and robots to interact effectively with their environment, particularly for tasks requiring the understanding of contact properties. Solving such dexterous manipulation tasks often relies on imitation learning from demonstration datasets, which are typically collect...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 15:51:18 GMT" } ]
2025-04-09T00:00:00
[ [ "Liu", "Fangchen", "" ], [ "Li", "Chuanyu", "" ], [ "Qin", "Yihua", "" ], [ "Shaw", "Ankit", "" ], [ "Xu", "Jing", "" ], [ "Abbeel", "Pieter", "" ], [ "Chen", "Rui", "" ] ]
TITLE: ViTaMIn: Learning Contact-Rich Tasks Through Robot-Free Visuo-Tactile Manipulation Interface ABSTRACT: Tactile information plays a crucial role for humans and robots to interact effectively with their environment, particularly for tasks requiring the understanding of contact properties. Solving such dexter...
2504.06158
Saad Wazir
Saad Wazir, Daeyoung Kim
Rethinking the Nested U-Net Approach: Enhancing Biomarker Segmentation with Attention Mechanisms and Multiscale Feature Fusion
Published in the Proceedings of the 2024 International Conference on Medical Imaging and Computer-Aided Diagnosis (MICAD 2024), Lecture Notes in Electrical Engineering (LNEE), Volume 1372, Springer Nature, Singapore
Lecture Notes in Electrical Engineering, vol. 1372, pp. 175-186, Springer Nature, Singapore, 2025
10.1007/978-981-96-3863-5_17
null
eess.IV cs.CV
http://creativecommons.org/licenses/by/4.0/
Identifying biomarkers in medical images is vital for a wide range of biotech applications. However, recent Transformer and CNN based methods often struggle with variations in morphology and staining, which limits their feature extraction capabilities. In medical image segmentation, where data samples are often limit...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 15:53:46 GMT" } ]
2025-04-09T00:00:00
[ [ "Wazir", "Saad", "" ], [ "Kim", "Daeyoung", "" ] ]
TITLE: Rethinking the Nested U-Net Approach: Enhancing Biomarker Segmentation with Attention Mechanisms and Multiscale Feature Fusion ABSTRACT: Identifying biomarkers in medical images is vital for a wide range of biotech applications. However, recent Transformer and CNN based methods often struggle with variatio...
2504.06166
Montgomery Gole
Montgomery Gole and Andriy Miranskyy
Assessing how hyperparameters impact Large Language Models' sarcasm detection performance
null
null
null
null
cs.CL
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sarcasm detection is challenging for both humans and machines. This work explores how model characteristics impact sarcasm detection in OpenAI's GPT, and Meta's Llama-2 models, given their strong natural language understanding, and popularity. We evaluate fine-tuned and zero-shot models across various sizes, releases...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 16:05:25 GMT" } ]
2025-04-09T00:00:00
[ [ "Gole", "Montgomery", "" ], [ "Miranskyy", "Andriy", "" ] ]
TITLE: Assessing how hyperparameters impact Large Language Models' sarcasm detection performance ABSTRACT: Sarcasm detection is challenging for both humans and machines. This work explores how model characteristics impact sarcasm detection in OpenAI's GPT, and Meta's Llama-2 models, given their strong natural lan...
2504.06176
Ian Groves
Ian Groves, Andrew Campbell, James Fernandes, Diego Rodriguez, Paul Murray, Massimiliano Vasile, Victoria Nockles
A Self-Supervised Framework for Space Object Behaviour Characterisation
15 pages, 10 figures
null
null
null
cs.LG cs.AI physics.space-ph
http://creativecommons.org/licenses/by-nc-nd/4.0/
Foundation Models, pre-trained on large unlabelled datasets before task-specific fine-tuning, are increasingly being applied to specialised domains. Recent examples include ClimaX for climate and Clay for satellite Earth observation, but a Foundation Model for Space Object Behavioural Analysis has not yet been develo...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 16:19:19 GMT" } ]
2025-04-09T00:00:00
[ [ "Groves", "Ian", "" ], [ "Campbell", "Andrew", "" ], [ "Fernandes", "James", "" ], [ "Rodriguez", "Diego", "" ], [ "Murray", "Paul", "" ], [ "Vasile", "Massimiliano", "" ], [ "Nockles", "Victoria", "" ] ]
TITLE: A Self-Supervised Framework for Space Object Behaviour Characterisation ABSTRACT: Foundation Models, pre-trained on large unlabelled datasets before task-specific fine-tuning, are increasingly being applied to specialised domains. Recent examples include ClimaX for climate and Clay for satellite Earth observ...
2504.06185
Vanessa Borst
Vanessa Borst, Timo Dittus, Tassilo Dege, Astrid Schmieder, and Samuel Kounev
WoundAmbit: Bridging State-of-the-Art Semantic Segmentation and Real-World Wound Care
Main paper: 17 pages; supplementary material: 16 pages; paper submitted to the application track of the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD 2025)
null
null
null
cs.CV cs.AI
http://creativecommons.org/licenses/by/4.0/
Chronic wounds affect a large population, particularly the elderly and diabetic patients, who often exhibit limited mobility and co-existing health conditions. Automated wound monitoring via mobile image capture can reduce in-person physician visits by enabling remote tracking of wound size. Semantic segmentation is ...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 16:25:59 GMT" } ]
2025-04-09T00:00:00
[ [ "Borst", "Vanessa", "" ], [ "Dittus", "Timo", "" ], [ "Dege", "Tassilo", "" ], [ "Schmieder", "Astrid", "" ], [ "Kounev", "Samuel", "" ] ]
TITLE: WoundAmbit: Bridging State-of-the-Art Semantic Segmentation and Real-World Wound Care ABSTRACT: Chronic wounds affect a large population, particularly the elderly and diabetic patients, who often exhibit limited mobility and co-existing health conditions. Automated wound monitoring via mobile image capture...
2504.06193
Zongyue Qin
Zongyue Qin, Shichang Zhang, Mingxuan Ju, Tong Zhao, Neil Shah, Yizhou Sun
Heuristic Methods are Good Teachers to Distill MLPs for Graph Link Prediction
null
null
null
null
cs.LG cs.AI
http://creativecommons.org/licenses/by/4.0/
Link prediction is a crucial graph-learning task with applications including citation prediction and product recommendation. Distilling Graph Neural Networks (GNNs) teachers into Multi-Layer Perceptrons (MLPs) students has emerged as an effective approach to achieve strong performance and reducing computational cost ...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 16:35:11 GMT" } ]
2025-04-09T00:00:00
[ [ "Qin", "Zongyue", "" ], [ "Zhang", "Shichang", "" ], [ "Ju", "Mingxuan", "" ], [ "Zhao", "Tong", "" ], [ "Shah", "Neil", "" ], [ "Sun", "Yizhou", "" ] ]
TITLE: Heuristic Methods are Good Teachers to Distill MLPs for Graph Link Prediction ABSTRACT: Link prediction is a crucial graph-learning task with applications including citation prediction and product recommendation. Distilling Graph Neural Networks (GNNs) teachers into Multi-Layer Perceptrons (MLPs) students ...
2504.06196
Shekoofeh Azizi
Eric Wang, Samuel Schmidgall, Paul F. Jaeger, Fan Zhang, Rory Pilgrim, Yossi Matias, Joelle Barral, David Fleet, Shekoofeh Azizi
TxGemma: Efficient and Agentic LLMs for Therapeutics
null
null
null
null
cs.AI cs.CL cs.LG
http://creativecommons.org/licenses/by/4.0/
Therapeutic development is a costly and high-risk endeavor that is often plagued by high failure rates. To address this, we introduce TxGemma, a suite of efficient, generalist large language models (LLMs) capable of therapeutic property prediction as well as interactive reasoning and explainability. Unlike task-speci...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 16:39:02 GMT" } ]
2025-04-09T00:00:00
[ [ "Wang", "Eric", "" ], [ "Schmidgall", "Samuel", "" ], [ "Jaeger", "Paul F.", "" ], [ "Zhang", "Fan", "" ], [ "Pilgrim", "Rory", "" ], [ "Matias", "Yossi", "" ], [ "Barral", "Joelle", "" ], [ "Fleet"...
TITLE: TxGemma: Efficient and Agentic LLMs for Therapeutics ABSTRACT: Therapeutic development is a costly and high-risk endeavor that is often plagued by high failure rates. To address this, we introduce TxGemma, a suite of efficient, generalist large language models (LLMs) capable of therapeutic property predictio...
2504.06207
Moncef Garouani
Moncef Garouani
An experimental survey and Perspective View on Meta-Learning for Automated Algorithms Selection and Parametrization
null
null
null
null
cs.LG cs.AI
http://creativecommons.org/licenses/by/4.0/
Considerable progress has been made in the recent literature studies to tackle the Algorithms Selection and Parametrization (ASP) problem, which is diversified in multiple meta-learning setups. Yet there is a lack of surveys and comparative evaluations that critically analyze, summarize and assess the performance of ...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 16:51:22 GMT" } ]
2025-04-09T00:00:00
[ [ "Garouani", "Moncef", "" ] ]
TITLE: An experimental survey and Perspective View on Meta-Learning for Automated Algorithms Selection and Parametrization ABSTRACT: Considerable progress has been made in the recent literature studies to tackle the Algorithms Selection and Parametrization (ASP) problem, which is diversified in multiple meta-lear...
2504.06219
Dongyang Fan
Dongyang Fan, Vinko Sabol\v{c}ec, Matin Ansaripour, Ayush Kumar Tarun, Martin Jaggi, Antoine Bosselut, Imanol Schlag
Can Performant LLMs Be Ethical? Quantifying the Impact of Web Crawling Opt-Outs
null
null
null
null
cs.CL cs.LG
http://creativecommons.org/licenses/by/4.0/
The increasing adoption of web crawling opt-outs by copyright holders of online content raises critical questions about the impact of data compliance on large language model (LLM) performance. However, little is known about how these restrictions (and the resultant filtering of pretraining datasets) affect the capabi...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 17:08:06 GMT" } ]
2025-04-09T00:00:00
[ [ "Fan", "Dongyang", "" ], [ "Sabolčec", "Vinko", "" ], [ "Ansaripour", "Matin", "" ], [ "Tarun", "Ayush Kumar", "" ], [ "Jaggi", "Martin", "" ], [ "Bosselut", "Antoine", "" ], [ "Schlag", "Imanol", "" ] ]
TITLE: Can Performant LLMs Be Ethical? Quantifying the Impact of Web Crawling Opt-Outs ABSTRACT: The increasing adoption of web crawling opt-outs by copyright holders of online content raises critical questions about the impact of data compliance on large language model (LLM) performance. However, little is known...
2504.06227
Krithi Shailya
Krithi Shailya, Shreya Rajpal, Gokul S Krishnan, Balaraman Ravindran
LExT: Towards Evaluating Trustworthiness of Natural Language Explanations
null
null
null
null
cs.CL
http://creativecommons.org/licenses/by/4.0/
As Large Language Models (LLMs) become increasingly integrated into high-stakes domains, there have been several approaches proposed toward generating natural language explanations. These explanations are crucial for enhancing the interpretability of a model, especially in sensitive domains like healthcare, where tra...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 17:16:52 GMT" } ]
2025-04-09T00:00:00
[ [ "Shailya", "Krithi", "" ], [ "Rajpal", "Shreya", "" ], [ "Krishnan", "Gokul S", "" ], [ "Ravindran", "Balaraman", "" ] ]
TITLE: LExT: Towards Evaluating Trustworthiness of Natural Language Explanations ABSTRACT: As Large Language Models (LLMs) become increasingly integrated into high-stakes domains, there have been several approaches proposed toward generating natural language explanations. These explanations are crucial for enhanc...
2504.06235
Shahryar Zehtabi
Shahryar Zehtabi, Dong-Jun Han, Seyyedali Hosseinalipour, Christopher G. Brinton
Decentralized Federated Domain Generalization with Style Sharing: A Formal Modeling and Convergence Analysis
null
null
null
null
cs.LG cs.AI
http://creativecommons.org/licenses/by/4.0/
Much of the federated learning (FL) literature focuses on settings where local dataset statistics remain the same between training and testing time. Recent advances in domain generalization (DG) aim to use data from source (training) domains to train a model that generalizes well to data from unseen target (testing) ...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 17:32:56 GMT" } ]
2025-04-09T00:00:00
[ [ "Zehtabi", "Shahryar", "" ], [ "Han", "Dong-Jun", "" ], [ "Hosseinalipour", "Seyyedali", "" ], [ "Brinton", "Christopher G.", "" ] ]
TITLE: Decentralized Federated Domain Generalization with Style Sharing: A Formal Modeling and Convergence Analysis ABSTRACT: Much of the federated learning (FL) literature focuses on settings where local dataset statistics remain the same between training and testing time. Recent advances in domain generalizatio...
2504.06237
Mina Bishay
Mina Bishay, Graham Page, Waleed Emad, and Mohammad Mavadati
Monitoring Viewer Attention During Online Ads
Presented at the ECCV 2024 Workshops
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Nowadays, video ads spread through numerous online platforms, and are being watched by millions of viewers worldwide. Big brands gauge the liking and purchase intent of their new ads, by analyzing the facial responses of viewers recruited online to watch the ads from home or work. Although this approach captures natu...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 17:34:02 GMT" } ]
2025-04-09T00:00:00
[ [ "Bishay", "Mina", "" ], [ "Page", "Graham", "" ], [ "Emad", "Waleed", "" ], [ "Mavadati", "Mohammad", "" ] ]
TITLE: Monitoring Viewer Attention During Online Ads ABSTRACT: Nowadays, video ads spread through numerous online platforms, and are being watched by millions of viewers worldwide. Big brands gauge the liking and purchase intent of their new ads, by analyzing the facial responses of viewers recruited online to watc...
2504.06263
Yiying Yang
Yiying Yang, Wei Cheng, Sijin Chen, Xianfang Zeng, Jiaxu Zhang, Liao Wang, Gang Yu, Xingjun Ma, Yu-Gang Jiang
OmniSVG: A Unified Scalable Vector Graphics Generation Model
18 pages; Project Page: https://omnisvg.github.io/
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Scalable Vector Graphics (SVG) is an important image format widely adopted in graphic design because of their resolution independence and editability. The study of generating high-quality SVG has continuously drawn attention from both designers and researchers in the AIGC community. However, existing methods either p...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 17:59:49 GMT" } ]
2025-04-09T00:00:00
[ [ "Yang", "Yiying", "" ], [ "Cheng", "Wei", "" ], [ "Chen", "Sijin", "" ], [ "Zeng", "Xianfang", "" ], [ "Zhang", "Jiaxu", "" ], [ "Wang", "Liao", "" ], [ "Yu", "Gang", "" ], [ "Ma", "Xingjun", ...
TITLE: OmniSVG: A Unified Scalable Vector Graphics Generation Model ABSTRACT: Scalable Vector Graphics (SVG) is an important image format widely adopted in graphic design because of their resolution independence and editability. The study of generating high-quality SVG has continuously drawn attention from both des...
2504.06264
Jisang Han
Jisang Han, Honggyu An, Jaewoo Jung, Takuya Narihira, Junyoung Seo, Kazumi Fukuda, Chaehyun Kim, Sunghwan Hong, Yuki Mitsufuji, Seungryong Kim
D^2USt3R: Enhancing 3D Reconstruction with 4D Pointmaps for Dynamic Scenes
project page: https://cvlab-kaist.github.io/DDUSt3R/
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
We address the task of 3D reconstruction in dynamic scenes, where object motions degrade the quality of previous 3D pointmap regression methods, such as DUSt3R, originally designed for static 3D scene reconstruction. Although these methods provide an elegant and powerful solution in static settings, they struggle in ...
[ { "version": "v1", "created": "Tue, 8 Apr 2025 17:59:50 GMT" } ]
2025-04-09T00:00:00
[ [ "Han", "Jisang", "" ], [ "An", "Honggyu", "" ], [ "Jung", "Jaewoo", "" ], [ "Narihira", "Takuya", "" ], [ "Seo", "Junyoung", "" ], [ "Fukuda", "Kazumi", "" ], [ "Kim", "Chaehyun", "" ], [ "Hong", ...
TITLE: D^2USt3R: Enhancing 3D Reconstruction with 4D Pointmaps for Dynamic Scenes ABSTRACT: We address the task of 3D reconstruction in dynamic scenes, where object motions degrade the quality of previous 3D pointmap regression methods, such as DUSt3R, originally designed for static 3D scene reconstruction. Altho...
2108.11328
Shibal Ibrahim
Shibal Ibrahim, Peter Radchenko, Emanuel Ben-David, Rahul Mazumder
Predicting Census Survey Response Rates With Parsimonious Additive Models and Structured Interactions
Published in Annals of Applied Statistics
The Annals of Applied Statistics 2025, Vol. 19, No. 1, 94-120
10.1214/24-AOAS1929
null
stat.ML cs.LG stat.AP stat.CO
http://creativecommons.org/licenses/by/4.0/
In this paper, we consider the problem of predicting survey response rates using a family of flexible and interpretable nonparametric models. The study is motivated by the US Census Bureau's well-known ROAM application, which uses a linear regression model trained on the US Census Planning Database data to identify h...
[ { "version": "v1", "created": "Tue, 24 Aug 2021 17:49:55 GMT" }, { "version": "v2", "created": "Wed, 8 Jun 2022 16:09:18 GMT" }, { "version": "v3", "created": "Fri, 26 May 2023 17:10:01 GMT" }, { "version": "v4", "created": "Thu, 7 Dec 2023 19:05:08 GMT" }, { "ver...
2025-04-08T00:00:00
[ [ "Ibrahim", "Shibal", "" ], [ "Radchenko", "Peter", "" ], [ "Ben-David", "Emanuel", "" ], [ "Mazumder", "Rahul", "" ] ]
TITLE: Predicting Census Survey Response Rates With Parsimonious Additive Models and Structured Interactions ABSTRACT: In this paper, we consider the problem of predicting survey response rates using a family of flexible and interpretable nonparametric models. The study is motivated by the US Census Bureau's well...
2201.12577
John Chiang
John Chiang
Volley Revolver: A Novel Matrix-Encoding Method for Privacy-Preserving Neural Networks (Inference)
The encoding method we proposed in this work, $\texttt{Volley Revolver}$, is particularly tailored for privacy-preserving neural networks. There is a great chance that it can be used to assist the private neural networks training, in which case for the backpropagation algorithm of the fully-connected layer the ...
null
null
null
cs.CR cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In this work, we present a novel matrix-encoding method that is particularly convenient for neural networks to make predictions in a privacy-preserving manner using homomorphic encryption. Based on this encoding method, we implement a convolutional neural network for handwritten image classification over encryption. ...
[ { "version": "v1", "created": "Sat, 29 Jan 2022 12:40:19 GMT" }, { "version": "v2", "created": "Sun, 14 Aug 2022 06:44:34 GMT" }, { "version": "v3", "created": "Wed, 29 Mar 2023 12:14:21 GMT" }, { "version": "v4", "created": "Tue, 9 Jan 2024 00:52:21 GMT" }, { "ve...
2025-04-08T00:00:00
[ [ "Chiang", "John", "" ] ]
TITLE: Volley Revolver: A Novel Matrix-Encoding Method for Privacy-Preserving Neural Networks (Inference) ABSTRACT: In this work, we present a novel matrix-encoding method that is particularly convenient for neural networks to make predictions in a privacy-preserving manner using homomorphic encryption. Based on ...
2305.12352
Wenzhi Gao
Yanguang Chen, Wenzhi Gao, Wanyu Zhang, Dongdong Ge, Huikang Liu, Yinyu Ye
Data-driven Mixed Integer Optimization through Probabilistic Multi-variable Branching
null
null
null
null
math.OC cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In this paper, we propose a Pre-trained Mixed Integer Optimization framework (PreMIO) that accelerates online mixed integer program (MIP) solving with offline datasets and machine learning models. Our method is based on a data-driven multi-variable cardinality branching procedure that splits the MIP feasible region u...
[ { "version": "v1", "created": "Sun, 21 May 2023 05:11:30 GMT" }, { "version": "v2", "created": "Tue, 5 Nov 2024 21:46:50 GMT" }, { "version": "v3", "created": "Fri, 4 Apr 2025 18:09:21 GMT" } ]
2025-04-08T00:00:00
[ [ "Chen", "Yanguang", "" ], [ "Gao", "Wenzhi", "" ], [ "Zhang", "Wanyu", "" ], [ "Ge", "Dongdong", "" ], [ "Liu", "Huikang", "" ], [ "Ye", "Yinyu", "" ] ]
TITLE: Data-driven Mixed Integer Optimization through Probabilistic Multi-variable Branching ABSTRACT: In this paper, we propose a Pre-trained Mixed Integer Optimization framework (PreMIO) that accelerates online mixed integer program (MIP) solving with offline datasets and machine learning models. Our method is ...
2307.00976
Ruitao Xie
Ruimin Ma, Ruitao Xie, Yanlin Wang, Jintao Meng, Yanjie Wei, Wenhui Xi, Yi Pan
Autism Spectrum Disorder Classification with Interpretability in Children based on Structural MRI Features Extracted using Contrastive Variational Autoencoder
null
Big Data Mining and Analytics, 2024, 7(3): 781-793
10.26599/BDMA.2024.9020004
null
cs.CV
http://creativecommons.org/licenses/by-nc-nd/4.0/
Autism spectrum disorder (ASD) is a highly disabling mental disease that brings significant impairments of social interaction ability to the patients, making early screening and intervention of ASD critical. With the development of the machine learning and neuroimaging technology, extensive research has been conducte...
[ { "version": "v1", "created": "Mon, 3 Jul 2023 12:46:19 GMT" }, { "version": "v2", "created": "Mon, 7 Apr 2025 08:32:48 GMT" } ]
2025-04-08T00:00:00
[ [ "Ma", "Ruimin", "" ], [ "Xie", "Ruitao", "" ], [ "Wang", "Yanlin", "" ], [ "Meng", "Jintao", "" ], [ "Wei", "Yanjie", "" ], [ "Xi", "Wenhui", "" ], [ "Pan", "Yi", "" ] ]
TITLE: Autism Spectrum Disorder Classification with Interpretability in Children based on Structural MRI Features Extracted using Contrastive Variational Autoencoder ABSTRACT: Autism spectrum disorder (ASD) is a highly disabling mental disease that brings significant impairments of social interaction ability to...
2307.14591
Junchao Huang
Junchao Huang, Xiaoqi He Yebo Wu and Sheng Zhao
The detection and rectification for identity-switch based on unfalsified control
null
null
null
null
cs.CV cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The purpose of multi-object tracking (MOT) is to continuously track and identify objects detected in videos. Currently, most methods for multi-object tracking model the motion information and combine it with appearance information to determine and track objects. In this paper, unfalsified control is employed to addre...
[ { "version": "v1", "created": "Thu, 27 Jul 2023 02:30:12 GMT" }, { "version": "v2", "created": "Sun, 6 Apr 2025 13:11:14 GMT" } ]
2025-04-08T00:00:00
[ [ "Huang", "Junchao", "" ], [ "Wu", "Xiaoqi He Yebo", "" ], [ "Zhao", "Sheng", "" ] ]
TITLE: The detection and rectification for identity-switch based on unfalsified control ABSTRACT: The purpose of multi-object tracking (MOT) is to continuously track and identify objects detected in videos. Currently, most methods for multi-object tracking model the motion information and combine it with appearan...
2307.16082
Mohammadali Sefidi Esfahani
Mohammadali Sefidi Esfahani, Mohammad Akbari
EnrichEvent: Enriching Social Data with Contextual Information for Emerging Event Extraction
null
null
null
null
cs.CL cs.AI
http://creativecommons.org/licenses/by-nc-sa/4.0/
Social platforms have emerged as crucial platforms for distributing information and discussing social events, offering researchers an excellent opportunity to design and implement novel event detection frameworks. Identifying unspecified events and detecting events without prior knowledge enables governments, aid age...
[ { "version": "v1", "created": "Sat, 29 Jul 2023 21:37:55 GMT" }, { "version": "v2", "created": "Wed, 16 Aug 2023 09:00:25 GMT" }, { "version": "v3", "created": "Mon, 25 Dec 2023 14:27:55 GMT" }, { "version": "v4", "created": "Wed, 27 Dec 2023 09:58:25 GMT" }, { "v...
2025-04-08T00:00:00
[ [ "Esfahani", "Mohammadali Sefidi", "" ], [ "Akbari", "Mohammad", "" ] ]
TITLE: EnrichEvent: Enriching Social Data with Contextual Information for Emerging Event Extraction ABSTRACT: Social platforms have emerged as crucial platforms for distributing information and discussing social events, offering researchers an excellent opportunity to design and implement novel event detection fr...
2309.02712
Amir H Gandomi
Shams Forruque Ahmed, Md. Sakib Bin Alam, Maliha Kabir, Shaila Afrin, Sabiha Jannat Rafa, Aanushka Mehjabin, Amir H. Gandomi
Unveiling the frontiers of deep learning: innovations shaping diverse domains
88 pages, 11 figures, 7 tables
Applied Intelligence, 55(7), 573 (2025)
10.1007/s10489-025-06259-x
null
cs.LG cs.AI cs.NE
http://creativecommons.org/licenses/by/4.0/
Deep learning (DL) allows computer models to learn, visualize, optimize, refine, and predict data. To understand its present state, examining the most recent advancements and applications of deep learning across various domains is essential. However, prior reviews focused on DL applications in only one or two domains...
[ { "version": "v1", "created": "Wed, 6 Sep 2023 04:50:39 GMT" }, { "version": "v2", "created": "Sat, 5 Apr 2025 01:29:03 GMT" } ]
2025-04-08T00:00:00
[ [ "Ahmed", "Shams Forruque", "" ], [ "Alam", "Md. Sakib Bin", "" ], [ "Kabir", "Maliha", "" ], [ "Afrin", "Shaila", "" ], [ "Rafa", "Sabiha Jannat", "" ], [ "Mehjabin", "Aanushka", "" ], [ "Gandomi", "Amir H.", ...
TITLE: Unveiling the frontiers of deep learning: innovations shaping diverse domains ABSTRACT: Deep learning (DL) allows computer models to learn, visualize, optimize, refine, and predict data. To understand its present state, examining the most recent advancements and applications of deep learning across various...
2309.14770
Haotian Li
Haotian Li, Bin Yu, Yuliang Wei, Kai Wang, Richard Yi Da Xu, Bailing Wang
KERMIT: Knowledge Graph Completion of Enhanced Relation Modeling with Inverse Transformation
Accepted to Knowledge-Based Systems
null
null
null
cs.CL
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Knowledge graph completion (KGC) revolves around populating missing triples in a knowledge graph using available information. Text-based methods, which depend on textual descriptions of triples, often encounter difficulties when these descriptions lack sufficient information for accurate prediction-an issue inherent ...
[ { "version": "v1", "created": "Tue, 26 Sep 2023 09:03:25 GMT" }, { "version": "v2", "created": "Sat, 3 Aug 2024 13:34:24 GMT" }, { "version": "v3", "created": "Mon, 7 Apr 2025 03:07:25 GMT" } ]
2025-04-08T00:00:00
[ [ "Li", "Haotian", "" ], [ "Yu", "Bin", "" ], [ "Wei", "Yuliang", "" ], [ "Wang", "Kai", "" ], [ "Da Xu", "Richard Yi", "" ], [ "Wang", "Bailing", "" ] ]
TITLE: KERMIT: Knowledge Graph Completion of Enhanced Relation Modeling with Inverse Transformation ABSTRACT: Knowledge graph completion (KGC) revolves around populating missing triples in a knowledge graph using available information. Text-based methods, which depend on textual descriptions of triples, often enc...
2310.08453
Jian Wu
Jian Wu, Carol Flannagan, Ulrich Sander, and Jonas B\"argman
Modeling Lead-vehicle Kinematics For Rear-end Crash Scenario Generation
null
IEEETrans.Intell.Transp.Syst. 25 (2024) 3176-3186
10.1109/TITS.2024.3369097
null
cs.RO
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The use of virtual safety assessment as the primary method for evaluating vehicle safety technologies has emphasized the importance of crash scenario generation. One of the most common crash types is the rear-end crash, which involves a lead vehicle and a following vehicle. Most studies have focused on the following ...
[ { "version": "v1", "created": "Fri, 22 Sep 2023 10:21:17 GMT" }, { "version": "v2", "created": "Fri, 13 Oct 2023 07:16:21 GMT" } ]
2025-04-08T00:00:00
[ [ "Wu", "Jian", "" ], [ "Flannagan", "Carol", "" ], [ "Sander", "Ulrich", "" ], [ "Bärgman", "Jonas", "" ] ]
TITLE: Modeling Lead-vehicle Kinematics For Rear-end Crash Scenario Generation ABSTRACT: The use of virtual safety assessment as the primary method for evaluating vehicle safety technologies has emphasized the importance of crash scenario generation. One of the most common crash types is the rear-end crash, which i...
2310.11439
Quentin Bouniot
Quentin Bouniot, Ievgen Redko, Anton Mallasto, Charlotte Laclau, Oliver Struckmeier, Karol Arndt, Markus Heinonen, Ville Kyrki, Samuel Kaski
From Alexnet to Transformers: Measuring the Non-linearity of Deep Neural Networks with Affine Optimal Transport
Code available at https://github.com/qbouniot/AffScoreDeep
null
null
null
cs.LG cs.AI stat.ML
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In the last decade, we have witnessed the introduction of several novel deep neural network (DNN) architectures exhibiting ever-increasing performance across diverse tasks. Explaining the upward trend of their performance, however, remains difficult as different DNN architectures of comparable depth and width -- comm...
[ { "version": "v1", "created": "Tue, 17 Oct 2023 17:50:22 GMT" }, { "version": "v2", "created": "Mon, 10 Jun 2024 09:29:21 GMT" }, { "version": "v3", "created": "Mon, 1 Jul 2024 14:39:54 GMT" }, { "version": "v4", "created": "Sun, 6 Apr 2025 16:31:38 GMT" } ]
2025-04-08T00:00:00
[ [ "Bouniot", "Quentin", "" ], [ "Redko", "Ievgen", "" ], [ "Mallasto", "Anton", "" ], [ "Laclau", "Charlotte", "" ], [ "Struckmeier", "Oliver", "" ], [ "Arndt", "Karol", "" ], [ "Heinonen", "Markus", "" ], ...
TITLE: From Alexnet to Transformers: Measuring the Non-linearity of Deep Neural Networks with Affine Optimal Transport ABSTRACT: In the last decade, we have witnessed the introduction of several novel deep neural network (DNN) architectures exhibiting ever-increasing performance across diverse tasks. Explaining t...
2310.14778
Jinzheng Zhao
Jinzheng Zhao, Yong Xu, Xinyuan Qian, Davide Berghi, Peipei Wu, Meng Cui, Jianyuan Sun, Philip J.B. Jackson and Wenwu Wang
Audio-Visual Speaker Tracking: Progress, Challenges, and Future Directions
null
null
null
null
cs.MM cs.SD eess.AS
http://creativecommons.org/licenses/by/4.0/
Audio-visual speaker tracking has drawn increasing attention over the past few years due to its academic values and wide applications. Audio and visual modalities can provide complementary information for localization and tracking. With audio and visual information, the Bayesian-based filter and deep learning-based m...
[ { "version": "v1", "created": "Mon, 23 Oct 2023 10:29:33 GMT" }, { "version": "v2", "created": "Sun, 17 Dec 2023 08:35:04 GMT" }, { "version": "v3", "created": "Thu, 19 Dec 2024 11:49:06 GMT" }, { "version": "v4", "created": "Sun, 6 Apr 2025 03:02:18 GMT" } ]
2025-04-08T00:00:00
[ [ "Zhao", "Jinzheng", "" ], [ "Xu", "Yong", "" ], [ "Qian", "Xinyuan", "" ], [ "Berghi", "Davide", "" ], [ "Wu", "Peipei", "" ], [ "Cui", "Meng", "" ], [ "Sun", "Jianyuan", "" ], [ "Jackson", "Phi...
TITLE: Audio-Visual Speaker Tracking: Progress, Challenges, and Future Directions ABSTRACT: Audio-visual speaker tracking has drawn increasing attention over the past few years due to its academic values and wide applications. Audio and visual modalities can provide complementary information for localization and ...
2310.18542
Shibal Ibrahim
Shibal Ibrahim and Kayhan Behdin and Rahul Mazumder
End-to-end Feature Selection Approach for Learning Skinny Trees
Published in AISTATS 2024
International Conference on Artificial Intelligence and Statistics (AISTATS) 2024
null
null
cs.LG
http://creativecommons.org/licenses/by/4.0/
We propose a new optimization-based approach for feature selection in tree ensembles, an important problem in statistics and machine learning. Popular tree ensemble toolkits e.g., Gradient Boosted Trees and Random Forests support feature selection post-training based on feature importance scores, while very popular, ...
[ { "version": "v1", "created": "Sat, 28 Oct 2023 00:15:10 GMT" }, { "version": "v2", "created": "Tue, 3 Sep 2024 07:34:54 GMT" }, { "version": "v3", "created": "Sun, 6 Apr 2025 03:10:53 GMT" } ]
2025-04-08T00:00:00
[ [ "Ibrahim", "Shibal", "" ], [ "Behdin", "Kayhan", "" ], [ "Mazumder", "Rahul", "" ] ]
TITLE: End-to-end Feature Selection Approach for Learning Skinny Trees ABSTRACT: We propose a new optimization-based approach for feature selection in tree ensembles, an important problem in statistics and machine learning. Popular tree ensemble toolkits e.g., Gradient Boosted Trees and Random Forests support featu...
2310.18651
Ali Javidani
Ali Javidani, Mohammad Amin Sadeghi, Babak Nadjar Araabi
Patch-Wise Self-Supervised Visual Representation Learning: A Fine-Grained Approach
15 pages
null
10.1007/s11760-025-04020-y
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Self-supervised visual representation learning traditionally focuses on image-level instance discrimination. Our study introduces an innovative, fine-grained dimension by integrating patch-level discrimination into these methodologies. This integration allows for the simultaneous analysis of local and global visual f...
[ { "version": "v1", "created": "Sat, 28 Oct 2023 09:35:30 GMT" }, { "version": "v2", "created": "Mon, 6 Nov 2023 07:52:31 GMT" }, { "version": "v3", "created": "Tue, 7 Nov 2023 07:02:59 GMT" }, { "version": "v4", "created": "Sat, 16 Dec 2023 10:50:45 GMT" }, { "ver...
2025-04-08T00:00:00
[ [ "Javidani", "Ali", "" ], [ "Sadeghi", "Mohammad Amin", "" ], [ "Araabi", "Babak Nadjar", "" ] ]
TITLE: Patch-Wise Self-Supervised Visual Representation Learning: A Fine-Grained Approach ABSTRACT: Self-supervised visual representation learning traditionally focuses on image-level instance discrimination. Our study introduces an innovative, fine-grained dimension by integrating patch-level discrimination into...
2311.00635
Andrea Giuseppe Di Francesco
Andrea Giuseppe Di Francesco, Giuliano Giampietro, Indro Spinelli and Danilo Comminiello
GATSY: Graph Attention Network for Music Artist Similarity
Accepted at IJCNN 2025
null
null
null
cs.IR
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The artist similarity quest has become a crucial subject in social and scientific contexts, driven by the desire to enhance music discovery according to user preferences. Modern research solutions facilitate music discovery according to user tastes. However, defining similarity among artists remains challenging due t...
[ { "version": "v1", "created": "Wed, 1 Nov 2023 16:36:19 GMT" }, { "version": "v2", "created": "Sat, 5 Apr 2025 18:14:41 GMT" } ]
2025-04-08T00:00:00
[ [ "Di Francesco", "Andrea Giuseppe", "" ], [ "Giampietro", "Giuliano", "" ], [ "Spinelli", "Indro", "" ], [ "Comminiello", "Danilo", "" ] ]
TITLE: GATSY: Graph Attention Network for Music Artist Similarity ABSTRACT: The artist similarity quest has become a crucial subject in social and scientific contexts, driven by the desire to enhance music discovery according to user preferences. Modern research solutions facilitate music discovery according to use...
2311.08176
Jingru Fu
Jingru Fu, Daniel Ferreira, \"Orjan Smedby, Rodrigo Moreno
A deformation-based morphometry framework for disentangling Alzheimer's disease from normal aging using learned normal aging templates
21 pages, 8 figures
null
10.1038/s41598-025-96234-w
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Alzheimer's Disease and normal aging are both characterized by brain atrophy. The question of whether AD-related brain atrophy represents accelerated aging or a neurodegeneration process distinct from that in normal aging remains unresolved. Moreover, precisely disentangling AD-related brain atrophy from normal aging...
[ { "version": "v1", "created": "Tue, 14 Nov 2023 14:04:35 GMT" } ]
2025-04-08T00:00:00
[ [ "Fu", "Jingru", "" ], [ "Ferreira", "Daniel", "" ], [ "Smedby", "Örjan", "" ], [ "Moreno", "Rodrigo", "" ] ]
TITLE: A deformation-based morphometry framework for disentangling Alzheimer's disease from normal aging using learned normal aging templates ABSTRACT: Alzheimer's Disease and normal aging are both characterized by brain atrophy. The question of whether AD-related brain atrophy represents accelerated aging or a n...
2312.00502
Aristotelis Ballas
Aristotelis Ballas, Vasileios Papapanagiotou and Christos Diou
Which Augmentation Should I Use? An Empirical Investigation of Augmentations for Self-Supervised Phonocardiogram Representation Learning
Accepted in IEEE ACCESS: https://doi.org/10.1109/ACCESS.2024.3519297
null
10.1109/ACCESS.2024.3519297
null
cs.LG cs.SD eess.AS q-bio.QM
http://creativecommons.org/licenses/by/4.0/
Despite recent advancements in deep learning, its application in real-world medical settings, such as phonocardiogram (PCG) classification, remains limited. A significant barrier is the lack of high-quality annotated datasets, which hampers the development of robust, generalizable models that can perform well on newl...
[ { "version": "v1", "created": "Fri, 1 Dec 2023 11:06:00 GMT" }, { "version": "v2", "created": "Mon, 18 Mar 2024 10:32:01 GMT" }, { "version": "v3", "created": "Fri, 5 Apr 2024 11:19:12 GMT" }, { "version": "v4", "created": "Wed, 11 Dec 2024 09:53:49 GMT" }, { "ver...
2025-04-08T00:00:00
[ [ "Ballas", "Aristotelis", "" ], [ "Papapanagiotou", "Vasileios", "" ], [ "Diou", "Christos", "" ] ]
TITLE: Which Augmentation Should I Use? An Empirical Investigation of Augmentations for Self-Supervised Phonocardiogram Representation Learning ABSTRACT: Despite recent advancements in deep learning, its application in real-world medical settings, such as phonocardiogram (PCG) classification, remains limited. A s...
2312.08034
Mushfiqur Rahman
Mushfiqur Rahman, Runze Liu, Chau-Wai Wong, Huaiyu Dai
Individualized Deepfake Detection Exploiting Traces Due to Double Neural-Network Operations
null
null
null
null
eess.IV cs.CR cs.CV cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In today's digital landscape, journalists urgently require tools to verify the authenticity of facial images and videos depicting specific public figures before incorporating them into news stories. Existing deepfake detectors are not optimized for this detection task when an image is associated with a specific and i...
[ { "version": "v1", "created": "Wed, 13 Dec 2023 10:21:00 GMT" }, { "version": "v2", "created": "Fri, 4 Apr 2025 21:05:01 GMT" } ]
2025-04-08T00:00:00
[ [ "Rahman", "Mushfiqur", "" ], [ "Liu", "Runze", "" ], [ "Wong", "Chau-Wai", "" ], [ "Dai", "Huaiyu", "" ] ]
TITLE: Individualized Deepfake Detection Exploiting Traces Due to Double Neural-Network Operations ABSTRACT: In today's digital landscape, journalists urgently require tools to verify the authenticity of facial images and videos depicting specific public figures before incorporating them into news stories. Existi...
2312.11952
Collin Leiber
Collin Leiber and Dominik Mautz and Claudia Plant and Christian B\"ohm
Automatic Parameter Selection for Non-Redundant Clustering
null
Proceedings of the 2022 SIAM International Conference on Data Mining (SDM) (pp. 226-234). Society for Industrial and Applied Mathematics
10.1137/1.9781611977172.26
null
cs.LG cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
High-dimensional datasets often contain multiple meaningful clusterings in different subspaces. For example, objects can be clustered either by color, weight, or size, revealing different interpretations of the given dataset. A variety of approaches are able to identify such non-redundant clusterings. However, most o...
[ { "version": "v1", "created": "Tue, 19 Dec 2023 08:53:00 GMT" }, { "version": "v2", "created": "Mon, 7 Apr 2025 07:13:36 GMT" } ]
2025-04-08T00:00:00
[ [ "Leiber", "Collin", "" ], [ "Mautz", "Dominik", "" ], [ "Plant", "Claudia", "" ], [ "Böhm", "Christian", "" ] ]
TITLE: Automatic Parameter Selection for Non-Redundant Clustering ABSTRACT: High-dimensional datasets often contain multiple meaningful clusterings in different subspaces. For example, objects can be clustered either by color, weight, or size, revealing different interpretations of the given dataset. A variety of a...
2401.07702
Christopher Davis
Christopher Davis, Andrew Caines, {\O}istein Andersen, Shiva Taslimipoor, Helen Yannakoudakis, Zheng Yuan, Christopher Bryant, Marek Rei, Paula Buttery
Prompting open-source and commercial language models for grammatical error correction of English learner text
8 pages with appendices; accepted to ACL Findings 2024
null
10.18653/v1/2024.findings-acl.711
null
cs.CL
http://creativecommons.org/licenses/by/4.0/
Thanks to recent advances in generative AI, we are able to prompt large language models (LLMs) to produce texts which are fluent and grammatical. In addition, it has been shown that we can elicit attempts at grammatical error correction (GEC) from LLMs when prompted with ungrammatical input sentences. We evaluate how...
[ { "version": "v1", "created": "Mon, 15 Jan 2024 14:19:47 GMT" }, { "version": "v2", "created": "Sun, 6 Apr 2025 11:25:39 GMT" } ]
2025-04-08T00:00:00
[ [ "Davis", "Christopher", "" ], [ "Caines", "Andrew", "" ], [ "Andersen", "Øistein", "" ], [ "Taslimipoor", "Shiva", "" ], [ "Yannakoudakis", "Helen", "" ], [ "Yuan", "Zheng", "" ], [ "Bryant", "Christopher", ...
TITLE: Prompting open-source and commercial language models for grammatical error correction of English learner text ABSTRACT: Thanks to recent advances in generative AI, we are able to prompt large language models (LLMs) to produce texts which are fluent and grammatical. In addition, it has been shown that we ca...
2401.09234
Alfredo Go\~ni Sarriguren
Alfredo Go\~ni Sarriguren
SARRIGUREN: a polynomial-time complete algorithm for random $k$-SAT with relatively dense clauses
24 pages, 2 figures, 8 tables, algorithms, results and data in https://goo.su/zV3Pt6E
null
null
null
cs.DS cs.CC
http://creativecommons.org/licenses/by-nc-sa/4.0/
SARRIGUREN, a new complete algorithm for SAT based on counting clauses (which is valid also for Unique-SAT and #SAT) is described, analyzed and tested. Although existing complete algorithms for SAT perform slower with clauses with many literals, that is an advantage for SARRIGUREN, because the more literals are in th...
[ { "version": "v1", "created": "Wed, 17 Jan 2024 14:23:55 GMT" }, { "version": "v2", "created": "Mon, 7 Apr 2025 08:42:46 GMT" } ]
2025-04-08T00:00:00
[ [ "Sarriguren", "Alfredo Goñi", "" ] ]
TITLE: SARRIGUREN: a polynomial-time complete algorithm for random $k$-SAT with relatively dense clauses ABSTRACT: SARRIGUREN, a new complete algorithm for SAT based on counting clauses (which is valid also for Unique-SAT and #SAT) is described, analyzed and tested. Although existing complete algorithms for SAT p...
2402.02085
Long Ma
Long Ma, Zhiyuan Yan, Qinglang Guo, Yong Liao, Haiyang Yu, Pengyuan Zhou
Detecting AI-Generated Video via Frame Consistency
null
null
null
null
cs.CV cs.AI
http://creativecommons.org/licenses/by/4.0/
The escalating quality of video generated by advanced video generation methods results in new security challenges, while there have been few relevant research efforts: 1) There is no open-source dataset for generated video detection, 2) No generated video detection method has been proposed so far. To this end, we pro...
[ { "version": "v1", "created": "Sat, 3 Feb 2024 08:52:06 GMT" }, { "version": "v2", "created": "Tue, 6 Feb 2024 02:51:00 GMT" }, { "version": "v3", "created": "Mon, 3 Jun 2024 11:00:25 GMT" }, { "version": "v4", "created": "Wed, 26 Jun 2024 03:32:50 GMT" }, { "vers...
2025-04-08T00:00:00
[ [ "Ma", "Long", "" ], [ "Yan", "Zhiyuan", "" ], [ "Guo", "Qinglang", "" ], [ "Liao", "Yong", "" ], [ "Yu", "Haiyang", "" ], [ "Zhou", "Pengyuan", "" ] ]
TITLE: Detecting AI-Generated Video via Frame Consistency ABSTRACT: The escalating quality of video generated by advanced video generation methods results in new security challenges, while there have been few relevant research efforts: 1) There is no open-source dataset for generated video detection, 2) No generate...
2402.05675
Tong Chen
Tong Chen, Raghavendra Selvan
Is Adversarial Training with Compressed Datasets Effective?
22 pages, 10 figures, 3 tables, accepted at Scandinavian Conference on Image Analysis 2025 (SCIA 2025)
null
null
null
cs.LG
http://creativecommons.org/licenses/by/4.0/
Dataset Condensation (DC) refers to the recent class of dataset compression methods that generate a smaller, synthetic, dataset from a larger dataset. This synthetic dataset aims to retain the essential information of the original dataset, enabling models trained on it to achieve performance levels comparable to thos...
[ { "version": "v1", "created": "Thu, 8 Feb 2024 13:53:11 GMT" }, { "version": "v2", "created": "Mon, 7 Apr 2025 17:31:31 GMT" } ]
2025-04-08T00:00:00
[ [ "Chen", "Tong", "" ], [ "Selvan", "Raghavendra", "" ] ]
TITLE: Is Adversarial Training with Compressed Datasets Effective? ABSTRACT: Dataset Condensation (DC) refers to the recent class of dataset compression methods that generate a smaller, synthetic, dataset from a larger dataset. This synthetic dataset aims to retain the essential information of the original dataset,...
2402.09081
Dan Garber
Dan Garber, Atara Kaplan
Low-Rank Extragradient Methods for Scalable Semidefinite Optimization
This version corrects an error in the previous version, as well as in the short version published in \textit{Operations Research Letters} \cite{garber2025low}: while in those versions we reported $\mathcal{O}(1/T)$ rates for the \textbf{best iterate}, in this corrected version these rates hold only w.r.t. the \...
null
null
null
math.OC cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We consider several classes of highly important semidefinite optimization problems that involve both a convex objective function (smooth or nonsmooth) and additional linear or nonlinear smooth and convex constraints, which are ubiquitous in statistics, machine learning, combinatorial optimization, and other domains. ...
[ { "version": "v1", "created": "Wed, 14 Feb 2024 10:48:00 GMT" }, { "version": "v2", "created": "Mon, 7 Apr 2025 09:36:31 GMT" } ]
2025-04-08T00:00:00
[ [ "Garber", "Dan", "" ], [ "Kaplan", "Atara", "" ] ]
TITLE: Low-Rank Extragradient Methods for Scalable Semidefinite Optimization ABSTRACT: We consider several classes of highly important semidefinite optimization problems that involve both a convex objective function (smooth or nonsmooth) and additional linear or nonlinear smooth and convex constraints, which are ub...
2402.14802
Andrea Giuseppe Di Francesco
Andrea Giuseppe Di Francesco, Francesco Caso, Maria Sofia Bucarelli and Fabrizio Silvestri
Link Prediction with Physics-Inspired Graph Neural Networks
Accepted at IJCNN 2025
null
null
null
cs.LG cs.IR cs.SI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The message-passing mechanism underlying Graph Neural Networks (GNNs) is not naturally suited for heterophilic datasets, where adjacent nodes often have different labels. Most solutions to this problem remain confined to the task of node classification. In this article, we focus on the valuable task of link predictio...
[ { "version": "v1", "created": "Thu, 22 Feb 2024 18:56:31 GMT" }, { "version": "v2", "created": "Sat, 5 Apr 2025 18:19:08 GMT" } ]
2025-04-08T00:00:00
[ [ "Di Francesco", "Andrea Giuseppe", "" ], [ "Caso", "Francesco", "" ], [ "Bucarelli", "Maria Sofia", "" ], [ "Silvestri", "Fabrizio", "" ] ]
TITLE: Link Prediction with Physics-Inspired Graph Neural Networks ABSTRACT: The message-passing mechanism underlying Graph Neural Networks (GNNs) is not naturally suited for heterophilic datasets, where adjacent nodes often have different labels. Most solutions to this problem remain confined to the task of node c...
2403.08462
Andrea Nini
Andrea Nini, Oren Halvani, Lukas Graner, Valerio Gherardi, Shunichi Ishihara
Grammar as a Behavioral Biometric: Using Cognitively Motivated Grammar Models for Authorship Verification
null
null
null
null
cs.CL cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Authorship Verification (AV) is a key area of research in digital text forensics, which addresses the fundamental question of whether two texts were written by the same person. Numerous computational approaches have been proposed over the last two decades in an attempt to address this challenge. However, existing AV ...
[ { "version": "v1", "created": "Wed, 13 Mar 2024 12:25:47 GMT" }, { "version": "v2", "created": "Mon, 7 Apr 2025 11:12:57 GMT" } ]
2025-04-08T00:00:00
[ [ "Nini", "Andrea", "" ], [ "Halvani", "Oren", "" ], [ "Graner", "Lukas", "" ], [ "Gherardi", "Valerio", "" ], [ "Ishihara", "Shunichi", "" ] ]
TITLE: Grammar as a Behavioral Biometric: Using Cognitively Motivated Grammar Models for Authorship Verification ABSTRACT: Authorship Verification (AV) is a key area of research in digital text forensics, which addresses the fundamental question of whether two texts were written by the same person. Numerous compu...
2403.10045
Eric Xue
Eric Xue, Yijiang Li, Haoyang Liu, Peiran Wang, Yifan Shen, Haohan Wang
Towards Adversarially Robust Dataset Distillation by Curvature Regularization
AAAI 2025
null
null
null
cs.LG cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Dataset distillation (DD) allows datasets to be distilled to fractions of their original size while preserving the rich distributional information, so that models trained on the distilled datasets can achieve a comparable accuracy while saving significant computational loads. Recent research in this area has been foc...
[ { "version": "v1", "created": "Fri, 15 Mar 2024 06:31:03 GMT" }, { "version": "v2", "created": "Thu, 19 Dec 2024 21:39:24 GMT" }, { "version": "v3", "created": "Mon, 31 Mar 2025 21:23:30 GMT" }, { "version": "v4", "created": "Fri, 4 Apr 2025 20:27:58 GMT" } ]
2025-04-08T00:00:00
[ [ "Xue", "Eric", "" ], [ "Li", "Yijiang", "" ], [ "Liu", "Haoyang", "" ], [ "Wang", "Peiran", "" ], [ "Shen", "Yifan", "" ], [ "Wang", "Haohan", "" ] ]
TITLE: Towards Adversarially Robust Dataset Distillation by Curvature Regularization ABSTRACT: Dataset distillation (DD) allows datasets to be distilled to fractions of their original size while preserving the rich distributional information, so that models trained on the distilled datasets can achieve a comparab...
2403.12529
Brian Godwin Lim
Brian Godwin Lim, Galvin Brice Sy Lim, Renzo Roel Tan, Kazushi Ikeda
Contextualized Messages Boost Graph Representations
Published in Transactions on Machine Learning Research
null
null
null
cs.LG
http://creativecommons.org/licenses/by/4.0/
Graph neural networks (GNNs) have gained significant attention in recent years for their ability to process data that may be represented as graphs. This has prompted several studies to explore their representational capability based on the graph isomorphism task. Notably, these works inherently assume a countable nod...
[ { "version": "v1", "created": "Tue, 19 Mar 2024 08:05:49 GMT" }, { "version": "v2", "created": "Wed, 22 May 2024 09:02:33 GMT" }, { "version": "v3", "created": "Mon, 30 Sep 2024 12:56:50 GMT" }, { "version": "v4", "created": "Mon, 7 Apr 2025 11:27:48 GMT" } ]
2025-04-08T00:00:00
[ [ "Lim", "Brian Godwin", "" ], [ "Lim", "Galvin Brice Sy", "" ], [ "Tan", "Renzo Roel", "" ], [ "Ikeda", "Kazushi", "" ] ]
TITLE: Contextualized Messages Boost Graph Representations ABSTRACT: Graph neural networks (GNNs) have gained significant attention in recent years for their ability to process data that may be represented as graphs. This has prompted several studies to explore their representational capability based on the graph i...
2403.15304
Yahya Badran
Yahya Badran, Christine Preisach
Addressing Label Leakage in Knowledge Tracing Models
null
Proceedings of the 17th International Conference on Computer Supported Education (CSEDU) - Volume 2, 2025, pp. 85-95
10.5220/0013275200003932
null
cs.CY cs.AI cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Knowledge Tracing (KT) is concerned with predicting students' future performance on learning items in intelligent tutoring systems. Learning items are tagged with skill labels called knowledge concepts (KCs). Many KT models expand the sequence of item-student interactions into KC-student interactions by replacing lea...
[ { "version": "v1", "created": "Fri, 22 Mar 2024 15:54:30 GMT" }, { "version": "v2", "created": "Thu, 11 Apr 2024 16:39:54 GMT" }, { "version": "v3", "created": "Mon, 7 Apr 2025 15:00:58 GMT" } ]
2025-04-08T00:00:00
[ [ "Badran", "Yahya", "" ], [ "Preisach", "Christine", "" ] ]
TITLE: Addressing Label Leakage in Knowledge Tracing Models ABSTRACT: Knowledge Tracing (KT) is concerned with predicting students' future performance on learning items in intelligent tutoring systems. Learning items are tagged with skill labels called knowledge concepts (KCs). Many KT models expand the sequence of...
2404.05014
Jinfa Huang
Shenghai Yuan, Jinfa Huang, Yujun Shi, Yongqi Xu, Ruijie Zhu, Bin Lin, Xinhua Cheng, Li Yuan, Jiebo Luo
MagicTime: Time-lapse Video Generation Models as Metamorphic Simulators
TPAMI 2025
null
null
null
cs.CV
http://creativecommons.org/licenses/by/4.0/
Recent advances in Text-to-Video generation (T2V) have achieved remarkable success in synthesizing high-quality general videos from textual descriptions. A largely overlooked problem in T2V is that existing models have not adequately encoded physical knowledge of the real world, thus generated videos tend to have lim...
[ { "version": "v1", "created": "Sun, 7 Apr 2024 16:49:07 GMT" }, { "version": "v2", "created": "Sun, 6 Apr 2025 03:43:41 GMT" } ]
2025-04-08T00:00:00
[ [ "Yuan", "Shenghai", "" ], [ "Huang", "Jinfa", "" ], [ "Shi", "Yujun", "" ], [ "Xu", "Yongqi", "" ], [ "Zhu", "Ruijie", "" ], [ "Lin", "Bin", "" ], [ "Cheng", "Xinhua", "" ], [ "Yuan", "Li", ...
TITLE: MagicTime: Time-lapse Video Generation Models as Metamorphic Simulators ABSTRACT: Recent advances in Text-to-Video generation (T2V) have achieved remarkable success in synthesizing high-quality general videos from textual descriptions. A largely overlooked problem in T2V is that existing models have not adeq...
2404.09654
Junran Wu
Jiaqi Zhu, Shaofeng Cai, Fang Deng, Beng Chin Ooi, Junran Wu
Do LLMs Understand Visual Anomalies? Uncovering LLM's Capabilities in Zero-shot Anomaly Detection
Accepted by MM'24 (Oral)
null
null
null
cs.CV cs.MM
http://creativecommons.org/licenses/by-nc-nd/4.0/
Large vision-language models (LVLMs) are markedly proficient in deriving visual representations guided by natural language. Recent explorations have utilized LVLMs to tackle zero-shot visual anomaly detection (VAD) challenges by pairing images with textual descriptions indicative of normal and abnormal conditions, re...
[ { "version": "v1", "created": "Mon, 15 Apr 2024 10:42:22 GMT" }, { "version": "v2", "created": "Tue, 10 Sep 2024 11:58:23 GMT" }, { "version": "v3", "created": "Mon, 7 Apr 2025 05:18:12 GMT" } ]
2025-04-08T00:00:00
[ [ "Zhu", "Jiaqi", "" ], [ "Cai", "Shaofeng", "" ], [ "Deng", "Fang", "" ], [ "Ooi", "Beng Chin", "" ], [ "Wu", "Junran", "" ] ]
TITLE: Do LLMs Understand Visual Anomalies? Uncovering LLM's Capabilities in Zero-shot Anomaly Detection ABSTRACT: Large vision-language models (LVLMs) are markedly proficient in deriving visual representations guided by natural language. Recent explorations have utilized LVLMs to tackle zero-shot visual anomaly ...
2404.13659
Tong Wang
Tong Wang, Guanzhou Chen, Xiaodong Zhang, Chenxi Liu, Xiaoliang Tan, Jiaqi Wang, Chanjuan He, Wenlin Zhou
LMFNet: An Efficient Multimodal Fusion Approach for Semantic Segmentation in High-Resolution Remote Sensing
null
null
10.1016/j.patcog.2025.111579
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Despite the rapid evolution of semantic segmentation for land cover classification in high-resolution remote sensing imagery, integrating multiple data modalities such as Digital Surface Model (DSM), RGB, and Near-infrared (NIR) remains a challenge. Current methods often process only two types of data, missing out on...
[ { "version": "v1", "created": "Sun, 21 Apr 2024 13:29:42 GMT" } ]
2025-04-08T00:00:00
[ [ "Wang", "Tong", "" ], [ "Chen", "Guanzhou", "" ], [ "Zhang", "Xiaodong", "" ], [ "Liu", "Chenxi", "" ], [ "Tan", "Xiaoliang", "" ], [ "Wang", "Jiaqi", "" ], [ "He", "Chanjuan", "" ], [ "Zhou", "...
TITLE: LMFNet: An Efficient Multimodal Fusion Approach for Semantic Segmentation in High-Resolution Remote Sensing ABSTRACT: Despite the rapid evolution of semantic segmentation for land cover classification in high-resolution remote sensing imagery, integrating multiple data modalities such as Digital Surface Mo...
2404.15451
Hongyi Cai
Hongyi Cai, Mohammad Mahdinur Rahman, Wenzhen Dong and Jingyu Wu
CFPFormer: Feature-pyramid like Transformer Decoder for Segmentation and Detection
null
null
null
null
cs.CV
http://creativecommons.org/licenses/by-nc-sa/4.0/
Feature pyramids have been widely adopted in convolutional neural networks and transformers for tasks in medical image segmentation. However, existing models generally focus on the Encoder-side Transformer for feature extraction. We further explore the potential in improving the feature decoder with a well-designed a...
[ { "version": "v1", "created": "Tue, 23 Apr 2024 18:46:07 GMT" }, { "version": "v2", "created": "Sat, 5 Apr 2025 23:18:49 GMT" } ]
2025-04-08T00:00:00
[ [ "Cai", "Hongyi", "" ], [ "Rahman", "Mohammad Mahdinur", "" ], [ "Dong", "Wenzhen", "" ], [ "Wu", "Jingyu", "" ] ]
TITLE: CFPFormer: Feature-pyramid like Transformer Decoder for Segmentation and Detection ABSTRACT: Feature pyramids have been widely adopted in convolutional neural networks and transformers for tasks in medical image segmentation. However, existing models generally focus on the Encoder-side Transformer for feat...
2405.00543
Kiet Nguyen
Quy Hoang Nguyen, Minh-Van Truong Nguyen, Kiet Van Nguyen
New Benchmark Dataset and Fine-Grained Cross-Modal Fusion Framework for Vietnamese Multimodal Aspect-Category Sentiment Analysis
null
Multimedia Systems 31, 4 (2025)
10.1007/s00530-024-01558-8
null
cs.CL cs.AI
http://creativecommons.org/licenses/by-nc-sa/4.0/
The emergence of multimodal data on social media platforms presents new opportunities to better understand user sentiments toward a given aspect. However, existing multimodal datasets for Aspect-Category Sentiment Analysis (ACSA) often focus on textual annotations, neglecting fine-grained information in images. Conse...
[ { "version": "v1", "created": "Wed, 1 May 2024 14:29:03 GMT" } ]
2025-04-08T00:00:00
[ [ "Nguyen", "Quy Hoang", "" ], [ "Nguyen", "Minh-Van Truong", "" ], [ "Van Nguyen", "Kiet", "" ] ]
TITLE: New Benchmark Dataset and Fine-Grained Cross-Modal Fusion Framework for Vietnamese Multimodal Aspect-Category Sentiment Analysis ABSTRACT: The emergence of multimodal data on social media platforms presents new opportunities to better understand user sentiments toward a given aspect. However, existing mult...
2405.04804
Yin Li
Yin Li, Rajalakshmi Nandakumar
WixUp: A General Data Augmentation Framework for Wireless Perception in Tracking of Humans
SenSys pre-published version
null
10.1145/3715014.3722084
null
cs.NI
http://creativecommons.org/licenses/by/4.0/
Recent advancements in wireless perception technologies, including mmWave, WiFi, and acoustics, have expanded their application in human motion tracking and health monitoring. They are promising alternatives to traditional camera-based perception systems, thanks to their efficacy under diverse conditions or occlusion...
[ { "version": "v1", "created": "Wed, 8 May 2024 04:26:32 GMT" }, { "version": "v2", "created": "Sat, 5 Apr 2025 20:25:46 GMT" } ]
2025-04-08T00:00:00
[ [ "Li", "Yin", "" ], [ "Nandakumar", "Rajalakshmi", "" ] ]
TITLE: WixUp: A General Data Augmentation Framework for Wireless Perception in Tracking of Humans ABSTRACT: Recent advancements in wireless perception technologies, including mmWave, WiFi, and acoustics, have expanded their application in human motion tracking and health monitoring. They are promising alternative...
2405.07765
Mubashara Akhtar
Mubashara Akhtar and Chenxi Pang and Andreea Marzoca and Yasemin Altun and Julian Martin Eisenschlos
TANQ: An open domain dataset of table answered questions
12 pages, accepted at TACL
null
null
null
cs.CL
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Language models, potentially augmented with tool usage such as retrieval are becoming the go-to means of answering questions. Understanding and answering questions in real-world settings often requires retrieving information from different sources, processing and aggregating data to extract insights, and presenting c...
[ { "version": "v1", "created": "Mon, 13 May 2024 14:07:20 GMT" }, { "version": "v2", "created": "Wed, 15 Jan 2025 07:29:20 GMT" }, { "version": "v3", "created": "Sat, 5 Apr 2025 10:44:55 GMT" } ]
2025-04-08T00:00:00
[ [ "Akhtar", "Mubashara", "" ], [ "Pang", "Chenxi", "" ], [ "Marzoca", "Andreea", "" ], [ "Altun", "Yasemin", "" ], [ "Eisenschlos", "Julian Martin", "" ] ]
TITLE: TANQ: An open domain dataset of table answered questions ABSTRACT: Language models, potentially augmented with tool usage such as retrieval are becoming the go-to means of answering questions. Understanding and answering questions in real-world settings often requires retrieving information from different so...
2405.07920
Ferdinand Schlatt
Ferdinand Schlatt, Maik Fr\"obe, Harrisen Scells, Shengyao Zhuang, Bevan Koopman, Guido Zuccon, Benno Stein, Martin Potthast, Matthias Hagen
Rank-DistiLLM: Closing the Effectiveness Gap Between Cross-Encoders and LLMs for Passage Re-Ranking
Accepted at ECIR'25
null
10.1007/978-3-031-88714-7_31
null
cs.IR
http://creativecommons.org/licenses/by/4.0/
Cross-encoders distilled from large language models (LLMs) are often more effective re-rankers than cross-encoders fine-tuned on manually labeled data. However, distilled models do not match the effectiveness of their teacher LLMs. We hypothesize that this effectiveness gap is due to the fact that previous work has n...
[ { "version": "v1", "created": "Mon, 13 May 2024 16:51:53 GMT" }, { "version": "v2", "created": "Sun, 16 Jun 2024 12:43:02 GMT" }, { "version": "v3", "created": "Sat, 22 Mar 2025 09:53:21 GMT" }, { "version": "v4", "created": "Sat, 5 Apr 2025 10:01:30 GMT" } ]
2025-04-08T00:00:00
[ [ "Schlatt", "Ferdinand", "" ], [ "Fröbe", "Maik", "" ], [ "Scells", "Harrisen", "" ], [ "Zhuang", "Shengyao", "" ], [ "Koopman", "Bevan", "" ], [ "Zuccon", "Guido", "" ], [ "Stein", "Benno", "" ], [ ...
TITLE: Rank-DistiLLM: Closing the Effectiveness Gap Between Cross-Encoders and LLMs for Passage Re-Ranking ABSTRACT: Cross-encoders distilled from large language models (LLMs) are often more effective re-rankers than cross-encoders fine-tuned on manually labeled data. However, distilled models do not match the ef...
2405.08487
Mian Zou
Mian Zou, Baosheng Yu, Yibing Zhan, Siwei Lyu, and Kede Ma
Semantic Contextualization of Face Forgery: A New Definition, Dataset, and Detection Method
null
null
null
null
cs.CV cs.CR
http://creativecommons.org/licenses/by/4.0/
In recent years, deep learning has greatly streamlined the process of manipulating photographic face images. Aware of the potential dangers, researchers have developed various tools to spot these counterfeits. Yet, none asks the fundamental question: What digital manipulations make a real photographic face image fake...
[ { "version": "v1", "created": "Tue, 14 May 2024 10:24:19 GMT" }, { "version": "v2", "created": "Sat, 29 Mar 2025 07:00:42 GMT" }, { "version": "v3", "created": "Sat, 5 Apr 2025 09:19:17 GMT" } ]
2025-04-08T00:00:00
[ [ "Zou", "Mian", "" ], [ "Yu", "Baosheng", "" ], [ "Zhan", "Yibing", "" ], [ "Lyu", "Siwei", "" ], [ "Ma", "Kede", "" ] ]
TITLE: Semantic Contextualization of Face Forgery: A New Definition, Dataset, and Detection Method ABSTRACT: In recent years, deep learning has greatly streamlined the process of manipulating photographic face images. Aware of the potential dangers, researchers have developed various tools to spot these counterfe...
2405.17238
Ziyang Li
Ziyang Li, Saikat Dutta, Mayur Naik
IRIS: LLM-Assisted Static Analysis for Detecting Security Vulnerabilities
null
null
null
null
cs.CR cs.PL cs.SE
http://creativecommons.org/licenses/by-nc-nd/4.0/
Software is prone to security vulnerabilities. Program analysis tools to detect them have limited effectiveness in practice due to their reliance on human labeled specifications. Large language models (or LLMs) have shown impressive code generation capabilities but they cannot do complex reasoning over code to detect...
[ { "version": "v1", "created": "Mon, 27 May 2024 14:53:35 GMT" }, { "version": "v2", "created": "Mon, 11 Nov 2024 21:05:43 GMT" }, { "version": "v3", "created": "Sun, 6 Apr 2025 23:46:59 GMT" } ]
2025-04-08T00:00:00
[ [ "Li", "Ziyang", "" ], [ "Dutta", "Saikat", "" ], [ "Naik", "Mayur", "" ] ]
TITLE: IRIS: LLM-Assisted Static Analysis for Detecting Security Vulnerabilities ABSTRACT: Software is prone to security vulnerabilities. Program analysis tools to detect them have limited effectiveness in practice due to their reliance on human labeled specifications. Large language models (or LLMs) have shown i...
2405.18902
Andrea Pugnana
Filippo Palomba and Andrea Pugnana and Jos\'e Manuel Alvarez and Salvatore Ruggieri
A Causal Framework for Evaluating Deferring Systems
Accepted at AISTATS 2025
null
null
null
cs.LG cs.AI stat.ML
http://creativecommons.org/licenses/by/4.0/
Deferring systems extend supervised Machine Learning (ML) models with the possibility to defer predictions to human experts. However, evaluating the impact of a deferring strategy on system accuracy is still an overlooked area. This paper fills this gap by evaluating deferring systems through a causal lens. We link t...
[ { "version": "v1", "created": "Wed, 29 May 2024 09:03:44 GMT" }, { "version": "v2", "created": "Mon, 7 Apr 2025 08:54:30 GMT" } ]
2025-04-08T00:00:00
[ [ "Palomba", "Filippo", "" ], [ "Pugnana", "Andrea", "" ], [ "Alvarez", "José Manuel", "" ], [ "Ruggieri", "Salvatore", "" ] ]
TITLE: A Causal Framework for Evaluating Deferring Systems ABSTRACT: Deferring systems extend supervised Machine Learning (ML) models with the possibility to defer predictions to human experts. However, evaluating the impact of a deferring strategy on system accuracy is still an overlooked area. This paper fills th...
2406.02541
Inkyu Shin
Inkyu Shin, Qihang Yu, Xiaohui Shen, In So Kweon, Kuk-Jin Yoon, Liang-Chieh Chen
Enhancing Temporal Consistency in Video Editing by Reconstructing Videos with 3D Gaussian Splatting
Accepted to TMLR 2025. Project page at https://video-3dgs-project.github.io/
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Recent advancements in zero-shot video diffusion models have shown promise for text-driven video editing, but challenges remain in achieving high temporal consistency. To address this, we introduce Video-3DGS, a 3D Gaussian Splatting (3DGS)-based video refiner designed to enhance temporal consistency in zero-shot vid...
[ { "version": "v1", "created": "Tue, 4 Jun 2024 17:57:37 GMT" }, { "version": "v2", "created": "Wed, 5 Jun 2024 05:00:39 GMT" }, { "version": "v3", "created": "Thu, 6 Jun 2024 01:40:56 GMT" }, { "version": "v4", "created": "Fri, 4 Apr 2025 18:48:54 GMT" } ]
2025-04-08T00:00:00
[ [ "Shin", "Inkyu", "" ], [ "Yu", "Qihang", "" ], [ "Shen", "Xiaohui", "" ], [ "Kweon", "In So", "" ], [ "Yoon", "Kuk-Jin", "" ], [ "Chen", "Liang-Chieh", "" ] ]
TITLE: Enhancing Temporal Consistency in Video Editing by Reconstructing Videos with 3D Gaussian Splatting ABSTRACT: Recent advancements in zero-shot video diffusion models have shown promise for text-driven video editing, but challenges remain in achieving high temporal consistency. To address this, we introduce...
2406.04928
Torben Peters
Ghjulia Sialelli, Torben Peters, Jan D. Wegner, Konrad Schindler
AGBD: A Global-scale Biomass Dataset
null
null
null
null
cs.CV cs.LG eess.IV
http://creativecommons.org/licenses/by-nc-sa/4.0/
Accurate estimates of Above Ground Biomass (AGB) are essential in addressing two of humanity's biggest challenges: climate change and biodiversity loss. Existing datasets for AGB estimation from satellite imagery are limited. Either they focus on specific, local regions at high resolution, or they offer global covera...
[ { "version": "v1", "created": "Fri, 7 Jun 2024 13:34:17 GMT" }, { "version": "v2", "created": "Mon, 9 Dec 2024 11:08:35 GMT" }, { "version": "v3", "created": "Mon, 7 Apr 2025 11:19:12 GMT" } ]
2025-04-08T00:00:00
[ [ "Sialelli", "Ghjulia", "" ], [ "Peters", "Torben", "" ], [ "Wegner", "Jan D.", "" ], [ "Schindler", "Konrad", "" ] ]
TITLE: AGBD: A Global-scale Biomass Dataset ABSTRACT: Accurate estimates of Above Ground Biomass (AGB) are essential in addressing two of humanity's biggest challenges: climate change and biodiversity loss. Existing datasets for AGB estimation from satellite imagery are limited. Either they focus on specific, local...
2406.09067
Tarun Khajuria
Tarun Khajuria, Braian Olmiro Dias, Marharyta Domnich, Jaan Aru
Interpreting the structure of multi-object representations in vision encoders
null
null
null
null
cs.CV cs.CL q-bio.NC
http://creativecommons.org/licenses/by/4.0/
In this work, we interpret the representations of multi-object scenes in vision encoders through the lens of structured representations. Structured representations allow modeling of individual objects distinctly and their flexible use based on the task context for both scene-level and object-specific tasks. These cap...
[ { "version": "v1", "created": "Thu, 13 Jun 2024 12:54:20 GMT" }, { "version": "v2", "created": "Tue, 18 Jun 2024 12:27:36 GMT" }, { "version": "v3", "created": "Sun, 6 Apr 2025 13:44:02 GMT" } ]
2025-04-08T00:00:00
[ [ "Khajuria", "Tarun", "" ], [ "Dias", "Braian Olmiro", "" ], [ "Domnich", "Marharyta", "" ], [ "Aru", "Jaan", "" ] ]
TITLE: Interpreting the structure of multi-object representations in vision encoders ABSTRACT: In this work, we interpret the representations of multi-object scenes in vision encoders through the lens of structured representations. Structured representations allow modeling of individual objects distinctly and the...
2406.09564
Ziyan Wang
Ziyan Wang, Xiaoming Huo, Hao Wang
Towards Domain Adaptive Neural Contextual Bandits
Accepted at ICLR 2025
null
null
null
cs.LG cs.AI cs.CE cs.CV stat.ML
http://creativecommons.org/licenses/by/4.0/
Contextual bandit algorithms are essential for solving real-world decision making problems. In practice, collecting a contextual bandit's feedback from different domains may involve different costs. For example, measuring drug reaction from mice (as a source domain) and humans (as a target domain). Unfortunately, ada...
[ { "version": "v1", "created": "Thu, 13 Jun 2024 20:12:46 GMT" }, { "version": "v2", "created": "Tue, 22 Oct 2024 02:14:24 GMT" }, { "version": "v3", "created": "Sun, 6 Apr 2025 18:23:33 GMT" } ]
2025-04-08T00:00:00
[ [ "Wang", "Ziyan", "" ], [ "Huo", "Xiaoming", "" ], [ "Wang", "Hao", "" ] ]
TITLE: Towards Domain Adaptive Neural Contextual Bandits ABSTRACT: Contextual bandit algorithms are essential for solving real-world decision making problems. In practice, collecting a contextual bandit's feedback from different domains may involve different costs. For example, measuring drug reaction from mice (as...
2406.19774
Yixing Li
Yixing Li, Yuxian Gu, Li Dong, Dequan Wang, Yu Cheng, Furu Wei
Direct Preference Knowledge Distillation for Large Language Models
null
null
null
null
cs.CL
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In the field of large language models (LLMs), Knowledge Distillation (KD) is a critical technique for transferring capabilities from teacher models to student models. However, existing KD methods face limitations and challenges in distillation of LLMs, including efficiency and insufficient measurement capabilities of...
[ { "version": "v1", "created": "Fri, 28 Jun 2024 09:23:40 GMT" }, { "version": "v2", "created": "Mon, 7 Apr 2025 06:11:54 GMT" } ]
2025-04-08T00:00:00
[ [ "Li", "Yixing", "" ], [ "Gu", "Yuxian", "" ], [ "Dong", "Li", "" ], [ "Wang", "Dequan", "" ], [ "Cheng", "Yu", "" ], [ "Wei", "Furu", "" ] ]
TITLE: Direct Preference Knowledge Distillation for Large Language Models ABSTRACT: In the field of large language models (LLMs), Knowledge Distillation (KD) is a critical technique for transferring capabilities from teacher models to student models. However, existing KD methods face limitations and challenges in d...
2407.00342
Kibeom Nam
Kibeom Nam
KPC-cF: Aspect-Based Sentiment Analysis via Implicit-Feature Alignment with Corpus Filtering
Work in Progress, DMLR@ICML 2024
null
null
null
cs.CL cs.AI
http://creativecommons.org/licenses/by-nc-sa/4.0/
Investigations into Aspect-Based Sentiment Analysis (ABSA) for Korean industrial reviews are notably lacking in the existing literature. Our research proposes an intuitive and effective framework for ABSA in low-resource languages such as Korean. It optimizes prediction labels by integrating translated benchmark and ...
[ { "version": "v1", "created": "Sat, 29 Jun 2024 07:01:51 GMT" }, { "version": "v2", "created": "Thu, 11 Jul 2024 17:08:36 GMT" }, { "version": "v3", "created": "Sat, 20 Jul 2024 09:32:01 GMT" }, { "version": "v4", "created": "Fri, 15 Nov 2024 17:59:10 GMT" }, { "v...
2025-04-08T00:00:00
[ [ "Nam", "Kibeom", "" ] ]
TITLE: KPC-cF: Aspect-Based Sentiment Analysis via Implicit-Feature Alignment with Corpus Filtering ABSTRACT: Investigations into Aspect-Based Sentiment Analysis (ABSA) for Korean industrial reviews are notably lacking in the existing literature. Our research proposes an intuitive and effective framework for ABSA...
2407.00923
Oleg Vasilyev
Oleg Vasilyev, Randy Sawaya, John Bohannon
Preserving Multilingual Quality While Tuning Query Encoder on English Only
Accepted to NAACL 2025
null
null
null
cs.CL
http://creativecommons.org/licenses/by/4.0/
A query encoder of a dual passage retrieval system can be tuned for specific types of queries or domains, while the precomputed and stored documents representations are kept intact. Switching from one query encoder to another when needed is easily feasible, unlike overhauling the embeddings of a whole knowledge base....
[ { "version": "v1", "created": "Mon, 1 Jul 2024 03:03:18 GMT" }, { "version": "v2", "created": "Fri, 9 Aug 2024 06:02:12 GMT" }, { "version": "v3", "created": "Sat, 14 Dec 2024 01:23:33 GMT" }, { "version": "v4", "created": "Sat, 5 Apr 2025 23:03:41 GMT" } ]
2025-04-08T00:00:00
[ [ "Vasilyev", "Oleg", "" ], [ "Sawaya", "Randy", "" ], [ "Bohannon", "John", "" ] ]
TITLE: Preserving Multilingual Quality While Tuning Query Encoder on English Only ABSTRACT: A query encoder of a dual passage retrieval system can be tuned for specific types of queries or domains, while the precomputed and stored documents representations are kept intact. Switching from one query encoder to anot...
2407.05952
Nikhil Abhyankar
Nikhil Abhyankar, Vivek Gupta, Dan Roth, Chandan K. Reddy
H-STAR: LLM-driven Hybrid SQL-Text Adaptive Reasoning on Tables
NAACL 2025 Main Conference
null
null
null
cs.DB cs.AI cs.CL cs.LG
http://creativecommons.org/licenses/by-sa/4.0/
Tabular reasoning involves interpreting natural language queries about tabular data, which presents a unique challenge of combining language understanding with structured data analysis. Existing methods employ either textual reasoning, which excels in semantic interpretation but struggles with mathematical operations...
[ { "version": "v1", "created": "Sat, 29 Jun 2024 21:24:19 GMT" }, { "version": "v2", "created": "Wed, 30 Oct 2024 23:44:31 GMT" }, { "version": "v3", "created": "Mon, 7 Apr 2025 00:44:34 GMT" } ]
2025-04-08T00:00:00
[ [ "Abhyankar", "Nikhil", "" ], [ "Gupta", "Vivek", "" ], [ "Roth", "Dan", "" ], [ "Reddy", "Chandan K.", "" ] ]
TITLE: H-STAR: LLM-driven Hybrid SQL-Text Adaptive Reasoning on Tables ABSTRACT: Tabular reasoning involves interpreting natural language queries about tabular data, which presents a unique challenge of combining language understanding with structured data analysis. Existing methods employ either textual reasoning,...
2407.13349
HongHao Li
Honghao Li, Yiwen Zhang, Yi Zhang, Hanwei Li, Lei Sang, and Jieming Zhu
FCN: Fusing Exponential and Linear Cross Network for Click-Through Rate Prediction
null
null
null
null
cs.IR
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
As an important modeling paradigm in click-through rate (CTR) prediction, the Deep & Cross Network (DCN) and its derivative models have gained widespread recognition primarily due to their success in a trade-off between computational cost and performance. This paradigm employs a cross network to explicitly model feat...
[ { "version": "v1", "created": "Thu, 18 Jul 2024 09:49:13 GMT" }, { "version": "v2", "created": "Fri, 19 Jul 2024 03:23:01 GMT" }, { "version": "v3", "created": "Mon, 29 Jul 2024 16:30:42 GMT" }, { "version": "v4", "created": "Wed, 31 Jul 2024 15:59:46 GMT" }, { "v...
2025-04-08T00:00:00
[ [ "Li", "Honghao", "" ], [ "Zhang", "Yiwen", "" ], [ "Zhang", "Yi", "" ], [ "Li", "Hanwei", "" ], [ "Sang", "Lei", "" ], [ "Zhu", "Jieming", "" ] ]
TITLE: FCN: Fusing Exponential and Linear Cross Network for Click-Through Rate Prediction ABSTRACT: As an important modeling paradigm in click-through rate (CTR) prediction, the Deep & Cross Network (DCN) and its derivative models have gained widespread recognition primarily due to their success in a trade-off be...
2407.19992
Hao Shu
Hao Shu
Enhancing Edge Detection by Texture Handling Architecture and Noiseless Training Data
28 pages
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Image edge detection (ED) is a fundamental task in computer vision. While convolution-based models have significantly advanced ED performance, achieving high precision under strict error tolerance constraints remains challenging. Furthermore, the reliance on noisy, human-annotated training data limits model performan...
[ { "version": "v1", "created": "Mon, 29 Jul 2024 13:24:55 GMT" }, { "version": "v2", "created": "Tue, 1 Oct 2024 12:22:31 GMT" }, { "version": "v3", "created": "Wed, 2 Oct 2024 10:24:45 GMT" }, { "version": "v4", "created": "Sat, 5 Apr 2025 02:17:03 GMT" } ]
2025-04-08T00:00:00
[ [ "Shu", "Hao", "" ] ]
TITLE: Enhancing Edge Detection by Texture Handling Architecture and Noiseless Training Data ABSTRACT: Image edge detection (ED) is a fundamental task in computer vision. While convolution-based models have significantly advanced ED performance, achieving high precision under strict error tolerance constraints re...
2408.07107
Wenxuan Yang
Wenxuan Yang, Hanyu Zhang, Weimin Tan, Yuqi Sun, Bo Yan
A Self-Supervised Paradigm for Data-Efficient Medical Foundation Model Pre-training: V-information Optimization Framework
null
null
null
null
cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Self-supervised pre-training medical foundation models on large-scale datasets demonstrate exceptional performance. Recent research challenges this common paradigm by introducing data-effective learning approaches, demonstrating that merely increasing pre-training data volume does not necessarily improve model perfor...
[ { "version": "v1", "created": "Tue, 13 Aug 2024 10:28:54 GMT" }, { "version": "v2", "created": "Fri, 16 Aug 2024 12:19:44 GMT" }, { "version": "v3", "created": "Sat, 23 Nov 2024 08:24:19 GMT" }, { "version": "v4", "created": "Sun, 6 Apr 2025 02:50:25 GMT" } ]
2025-04-08T00:00:00
[ [ "Yang", "Wenxuan", "" ], [ "Zhang", "Hanyu", "" ], [ "Tan", "Weimin", "" ], [ "Sun", "Yuqi", "" ], [ "Yan", "Bo", "" ] ]
TITLE: A Self-Supervised Paradigm for Data-Efficient Medical Foundation Model Pre-training: V-information Optimization Framework ABSTRACT: Self-supervised pre-training medical foundation models on large-scale datasets demonstrate exceptional performance. Recent research challenges this common paradigm by introduc...
2408.07108
Marina Zajnulina
Marina Zajnulina
Shannon Entropy Helps Optimize the Performance of a Frequency-Multiplexed Extreme Learning Machine
null
null
null
null
physics.optics
http://creativecommons.org/licenses/by/4.0/
Knowing the dynamics of neuromorphic photonic schemes would allow their optimization for controlled data-processing capability at possibly minimized energy consumption levels. In nonlinear substrates such as optical fibers or semiconductors, these dynamics can widely vary depending on each optical input encoded with ...
[ { "version": "v1", "created": "Tue, 13 Aug 2024 11:03:54 GMT" }, { "version": "v2", "created": "Sun, 6 Apr 2025 22:21:54 GMT" } ]
2025-04-08T00:00:00
[ [ "Zajnulina", "Marina", "" ] ]
TITLE: Shannon Entropy Helps Optimize the Performance of a Frequency-Multiplexed Extreme Learning Machine ABSTRACT: Knowing the dynamics of neuromorphic photonic schemes would allow their optimization for controlled data-processing capability at possibly minimized energy consumption levels. In nonlinear substrate...