arXivDaily arXiv每日学术速递 周一至周五更新

期刊&会议

International Conference on Learning Representations · 会议 · Machine Learning

共收录 9454
2503.18225 2025-05-20 cs.LG cs.CL cs.CV

DeLoRA: Decoupling Angles and Strength in Low-rank Adaptation

Massimo Bini, Leander Girrbach, Zeynep Akata

机构 * University of Tübingen(图宾根大学) Tübingen AI Center(图宾根人工智能中心) Helmholtz Munich(海德堡-穆恩医疗中心) Technical University of Munich(慕尼黑技术大学) Munich Center for Machine Learning(慕尼黑机器学习中心) MDSI(慕尼黑数据科学研究所)

Comments ICLR 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2502.12944 2025-05-20 cs.LG

Performance of Zero-Shot Time Series Foundation Models on Cloud Data

William Toner, Thomas L. Lee, Artjom Joosen, Rajkarn Singh, Martin Asenov

机构 * Systems Infrastructure Lab, Edinburgh Research Centre, Huawei(系统基础设施实验室,爱丁堡研究中心,华为)

Comments 5 pages, presented at the "I Can't Believe It's Not Better" workshop at ICLR 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2502.07830 2025-05-20 cs.CV cs.AI cs.LG

Captured by Captions: On Memorization and its Mitigation in CLIP Models

Wenhao Wang, Adam Dziedzic, Grace C. Kim, Michael Backes, Franziska Boenisch

机构 * CISPA Georgia Institute of Technology(佐治亚理工学院)

Comments Accepted at ICLR 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2411.19722 2025-05-20 cs.LG cs.AI cs.CV

JetFormer: An Autoregressive Generative Model of Raw Images and Text

Michael Tschannen, André Susano Pinto, Alexander Kolesnikov

机构 * Google DeepMind(谷歌DeepMind)

Comments ICLR 2025. Code available at https://github.com/google-research/big_vision

详情

展开后加载摘要…

URL PDF HTML 收藏
2306.03985 2025-05-20 cs.LG

Agent Performing Autonomous Stock Trading under Good and Bad Situations

Yunfei Luo, Zhangqi Duan

机构 * Manning College of Information and Computer Science University of Massachusetts Amherst(信息与计算机科学曼宁学院马萨诸塞大学阿姆赫斯特分校)

Comments Published as a workshop paper at ICLR 2023: AI for Agent Based Modeling

详情

展开后加载摘要…

URL PDF HTML 收藏
2503.18258 2025-05-20 cs.LG cs.AI

Severing Spurious Correlations with Data Pruning

Varun Mulchandani, Jung-Eun Kim

机构 * Department of Computer Science(计算机科学系) North Carolina State University(北卡罗来纳州立大学)

Comments ICLR 2025, Spotlight

详情

展开后加载摘要…

URL PDF HTML 收藏
2503.14338 2025-05-20 cs.LG

Higher-Order Graphon Neural Networks: Approximation and Cut Distance

Daniel Herbst, Stefanie Jegelka

机构 * TUM, School of CIT(图宁大学,信息与通信技术学院) TUM, MCML and MDSI(图宁大学,MCML和MDSI) MIT, Department of EECS and CSAIL(麻省理工学院,电子工程与计算机科学系和计算机科学与人工智能实验室)

Comments 53 pages, 6 figures, 2 tables. ICLR 2025 camera ready

详情

展开后加载摘要…

URL PDF HTML 收藏
2503.09573 2025-05-20 cs.LG cs.AI

Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models

Marianne Arriola, Aaron Gokaslan, Justin T. Chiu, Zhihan Yang, Zhixuan Qi, Jiaqi Han, Subham Sekhar Sahoo, Volodymyr Kuleshov

Comments ICLR 2025 Oral. We provide the code at https://github.com/kuleshov-group/bd3lms

详情

展开后加载摘要…

URL PDF HTML 收藏
2503.09543 2025-05-20 cs.CL cs.LG

PolyPythias: Stability and Outliers across Fifty Language Model Pre-Training Runs

Oskar van der Wal, Pietro Lesci, Max Muller-Eberstein, Naomi Saphra, Hailey Schoelkopf, Willem Zuidema, Stella Biderman

机构 * University of Amsterdam(阿姆斯特丹大学) University of Cambridge(剑桥大学) IT University of Copenhagen(哥本哈根技术大学) Harvard University(哈佛大学) Anthropic(Anthropic公司) EleutherAI

Comments Published as a conference paper at ICLR 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2503.00856 2025-05-20 stat.ML cs.LG

Asymptotic Analysis of Two-Layer Neural Networks after One Gradient Step under Gaussian Mixtures Data with Structure

Samet Demir, Zafer Dogan

机构 * MLIP Research Group, KUIS AI Center, Koç University(MLIP研究组、KUIS人工智能中心、科举大学) Department of EEE, Koç University(电子工程系、科举大学)

Comments ICLR 2025, 27 pages, 9 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2502.03950 2025-05-20 cs.CV

LR0.FM: Low-Res Benchmark and Improving Robustness for Zero-Shot Classification in Foundation Models

Priyank Pathak, Shyam Marjit, Shruti Vyas, Yogesh S Rawat

机构 * University of Central Florida(佛罗里达中央大学) IIIT Guwahati(古瓦哈提理工学院)

Comments Accepted to ICLR 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2502.03052 2025-05-20 cs.LG cs.CR

Understanding and Enhancing the Transferability of Jailbreaking Attacks

Runqi Lin, Bo Han, Fengwang Li, Tongling Liu

机构 * Sydney AI Centre, The University of Sydney(悉尼人工智能中心、悉尼大学) Hong Kong Baptist University(香港 Baptist 大学)

Comments Accepted by ICLR 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2412.09945 2025-05-20 cs.CV

Going Beyond Feature Similarity: Effective Dataset Distillation based on Class-Aware Conditional Mutual Information

Xinhao Zhong, Bin Chen, Hao Fang, Xulin Gu, Shu-Tao Xia, En-Hui Yang

机构 * Harbin Institute of Technology, Shenzhen(哈尔滨工业大学(深圳)) Peng Cheng Laboratory(鹏城实验室) Tsinghua Shenzhen International Graduate School, Tsinghua University(清华大学深圳国际研究生院) University of Waterloo(滑铁卢大学)

Comments Accepted to ICLR 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2412.07775 2025-05-20 cs.LG cs.CV

Efficient Diversity-Preserving Diffusion Alignment via Gradient-Informed GFlowNets

Zhen Liu, Tim Z. Xiao, Weiyang Liu, Yoshua Bengio, Dinghuai Zhang

机构 * Mila, Université de Montréal(蒙特利尔大学Mila实验室) Max Planck Institute for Intelligent Systems - Tübingen(智能系统马克斯·普朗克研究所(图宾根)) The Chinese University of Hong Kong (Shenzhen)(香港中文大学(深圳)) University of Tübingen(图宾根大学) University of Cambridge(剑桥大学) Microsoft Research(微软研究院)

Comments Technical Report (36 pages, 31 figures), Accepted at ICLR 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2410.16032 2025-05-20 cs.LG cs.AI

TimeMixer++: A General Time Series Pattern Machine for Universal Predictive Analysis

Shiyu Wang, Jiawei Li, Xiaoming Shi, Zhou Ye, Baichuan Mo, Wenze Lin, Shengtong Ju, Zhixuan Chu, Ming Jin

机构 * Griffith University(格里菲斯大学) The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州)) Massachusetts Institute of Technology(麻省理工学院) Zhejiang University(浙江大学) The State Key Laboratory of Blockchain and Data Security(区块链与数据安全国家重点实验室) Hangzhou High-Tech Zone (Binjiang) Institute of Blockchain and Data Security(杭州高新技术区(滨江)区块链与数据安全研究院)

Comments Accepted by the 13th International Conference on Learning Representations (ICLR 2025)

详情

展开后加载摘要…

URL PDF HTML 收藏
2410.10733 2025-05-20 cs.CV cs.AI

Deep Compression Autoencoder for Efficient High-Resolution Diffusion Models

Junyu Chen, Han Cai, Junsong Chen, Enze Xie, Shang Yang, Haotian Tang, Muyang Li, Yao Lu, Song Han

机构 * MIT(麻省理工学院) Tsinghua University(清华大学) NVIDIA(英伟达)

Comments ICLR 2025. The first two authors contributed equally to this work. Fix Typo

详情

展开后加载摘要…

URL PDF HTML 收藏
2410.06577 2025-05-20 cs.CL

Rodimus*: Breaking the Accuracy-Efficiency Trade-Off with Efficient Attentions

Zhihao He, Hang Yu, Zi Gong, Shizhan Liu, Jianguo Li, Weiyao Lin

机构 * Shanghai Jiao Tong University(上海交通大学) Ant Group(蚂蚁集团)

Comments Accepted by ICLR 2025. Camera-ready Version

详情

展开后加载摘要…

URL PDF HTML 收藏
2410.03968 2025-05-20 cs.LG cs.AI cs.GT math.OC

Decoding Game: On Minimax Optimality of Heuristic Text Generation Strategies

Sijin Chen, Omar Hagrass, Jason M. Klusowski

机构 * Department of Electrical and Computer Engineering, Princeton University(普林斯顿大学电气工程与计算机科学系) Department of Operations Research and Financial Engineering, Princeton University(普林斯顿大学运筹学与金融工程系)

Comments 20 pages, accepted to ICLR 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2410.00645 2025-05-20 cs.LG

LoRanPAC: Low-rank Random Features and Pre-trained Models for Bridging Theory and Practice in Continual Learning

Liangzu Peng, Juan Elenter, Joshua Agterberg, Alejandro Ribeiro, René Vidal

机构 * University of Pennsylvania(宾夕法尼亚大学) University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校)

Comments 47 pages, 18 figures, 16 tables (v3, accepted to ICLR 2025)

详情

展开后加载摘要…

URL PDF HTML 收藏
2409.05657 2025-05-20 cs.LG

Adversarial Attacks on Data Attribution

Xinhe Wang, Pingbang Hu, Junwei Deng, Jiaqi W. Ma

机构 * University of Michigan(密歇根大学) University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校)

Comments Accepted at the 13th International Conference on Learning Representations (ICLR 2025)

详情

展开后加载摘要…

URL PDF HTML 收藏
2407.02772 2025-05-20 cs.LG cs.CL cs.CV

Gradient descent with generalized Newton's method

Zhiqi Bu, Shiyun Xu

机构 * University of Pennsylvania(宾夕法尼亚大学)

Comments Accepted to ICLR 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2406.00434 2025-05-20 cs.CV

MoDGS: Dynamic Gaussian Splatting from Casually-captured Monocular Videos with Depth Priors

Qingming Liu, Yuan Liu, Jiepeng Wang, Xianqiang Lyv, Peng Wang, Wenping Wang, Junhui Hou

机构 * City University of HongKong(香港城市大学) HKUST(香港科技大学) HKU(香港大学) TAMU(德克萨斯大学奥斯汀分校) CUHK(SZ)(香港城市大学(深圳))

Comments Accepted as a poster at ICLR. Project page: https://modgs.github.io

详情

展开后加载摘要…

URL PDF HTML 收藏
2305.15560 2025-05-20 cs.CV cs.CR cs.LG

Differentially Private Synthetic Data via Foundation Model APIs 1: Images

Zinan Lin, Sivakanth Gopi, Janardhan Kulkarni, Harsha Nori, Sergey Yekhanin

机构 * Microsoft Research(微软研究院)

Comments Published in ICLR 2024

详情

展开后加载摘要…

URL PDF HTML 收藏
2410.15433 2025-05-19 q-bio.NC cs.CV cs.LG stat.ML

Discriminating image representations with principal distortions

Jenelle Feather, David Lipshutz, Sarah E. Harvey, Alex H. Williams, Eero P. Simoncelli

机构 * Center for Computational Neuroscience, Flatiron Institute, Simons Foundation(计算神经科学中心,Flatiron研究所,Simons基金会) Department of Neuroscience, Baylor College of Medicine(神经科学系,贝勒医学院) Center for Neural Science, New York University(神经科学中心,纽约大学)

Journal ref Int'l Conf on Learning Representations (ICLR), vol 13, Singapore, May 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2410.08893 2025-05-19 cs.LG cs.AI cs.RO

Drama: Mamba-Enabled Model-Based Reinforcement Learning Is Sample and Parameter Efficient

Wenlong Wang, Ivana Dusparic, Yucheng Shi, Ke Zhang, Vinny Cahill

机构 * School of Computer Science and Statistics(计算机科学与统计学系)

Comments Published as a conference paper at ICLR 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2407.06325 2025-05-19 cs.LG cs.DC math.OC

CONGO: Compressive Online Gradient Optimization

Jeremy Carleton, Prathik Vijaykumar, Divyanshu Saxena, Dheeraj Narasimha, Srinivas Shakkottai, Aditya Akella

机构 * Texas A&M University(德克萨斯A&M大学) The University of Texas at Austin(德克萨斯大学奥斯汀分校) Inria(法国国家信息与自动化研究所)

Comments Accepted at ICLR 2025; 34 pages, 12 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2406.11624 2025-05-19 cs.LG cs.CL cs.CV

Words in Motion: Extracting Interpretable Control Vectors for Motion Transformers

Omer Sahin Tas, Royden Wagner

Comments ICLR 2025 final version. Our implementation is available at https://github.com/kit-mrt/future-motion

详情

展开后加载摘要…

URL PDF HTML 收藏
2505.11245 2025-05-19 cs.CV

Diffusion-NPO: Negative Preference Optimization for Better Preference Aligned Generation of Diffusion Models

Fu-Yun Wang, Yunhao Shui, Jingtan Piao, Keqiang Sun, Hongsheng Li

机构 * MMLab, CUHK, Hong Kong(CUHK的MMLab, 香港) Shanghai Jiang Tong University, Shanghai(上海江 Tong大学, 上海) CPII under InnoHK, Hong Kong(InnoHK下的CPII, 香港)

Comments Accepted to ICLR 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2505.10941 2025-05-19 cs.LG

Privacy-Aware Lifelong Learning

Ozan Özdenizci, Elmar Rueckert, Robert Legenstein

机构 * Chair of Cyber-Physical-Systems(网络物理系统主任) Institute of Machine Learning and Neural Computation(机器学习与神经计算研究所)

Journal ref International Conference on Learning Representations (ICLR) 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2410.11758 2025-05-16 cs.RO cs.CL cs.CV cs.LG

Latent Action Pretraining from Videos

Seonghyeon Ye, Joel Jang, Byeongguk Jeon, Sejune Joo, Jianwei Yang, Baolin Peng, Ajay Mandlekar, Reuben Tan, Yu-Wei Chao, Bill Yuchen Lin, Lars Liden, Kimin Lee, Jianfeng Gao, Luke Zettlemoyer, Dieter Fox, Minjoon Seo

机构 * KAIST(韩国科学技术院) University of Washington(华盛顿大学) Microsoft Research(微软研究院) NVIDIA(英伟达) Allen Institute for AI(人工智能算法研究所)

Comments ICLR 2025 Website: https://latentactionpretraining.github.io

详情

展开后加载摘要…

URL PDF HTML 收藏