ICML

Deep Learning on Real-World Graphs

Emanuele Rossi

Comments The thesis was submitted for the degree of Doctor of Philosophy in Computing at Imperial College London (February 2024), under the supervision of Prof. Michael M. Bronstein. It includes work published at ICML, ICLR, NeurIPS, and the Learning on Graphs Conference

Calibrated Language Models and How to Find Them with Label Smoothing

Jerry Huang, Peng Lu, Qiuhao Zeng

Comments Accepted to the Forty-second International Conference on Machine Learning (ICML) 2025. First two authors contributed equally. Official proceedings version available at https://proceedings.mlr.press/v267/huang25w.html

Proactive Agents for Multi-Turn Text-to-Image Generation Under Uncertainty

Meera Hahn, Wenjun Zeng, Nithish Kannen, Rich Galt, Kartikeya Badola, Been Kim, Zi Wang

机构 * Google DeepMind（谷歌DeepMind） 纠错

Journal ref International Conference on Machine Learning, 2025

Fair Clustering via Alignment

Kunwoong Kim, Jihu Lee, Sangchul Park, Yongdai Kim

机构 * Department of Statistics, Seoul National University, Republic of Korea（统计系，首尔国立大学，大韩民国） ； School of Law, Seoul National University, Republic of Korea（法学院，首尔国立大学，大韩民国） 纠错

Journal ref ICML 2025 (Forty-Second International Conference on Machine Learning)

Gatekeeper: Improving Model Cascades Through Confidence Tuning

Stephan Rabanser, Nathalie Rauschmayr, Achin Kulshrestha, Petra Poklukar, Wittawat Jitkrittum, Sean Augenstein, Congchao Wang, Federico Tombari

机构 * Princeton University（普林斯顿大学） ； Google（谷歌） 纠错

Comments Presented at the TTODLer-FM workshop at the International Conference on Machine Learning (ICML) 2025

Log Neural Controlled Differential Equations: The Lie Brackets Make a Difference

Benjamin Walker, Andrew D. McLeod, Tiexin Qin, Yichuan Cheng, Haoliang Li, Terry Lyons

机构 * Mathematical Institute, University of Oxford, UK（牛津大学数学研究所） ； Department of Electrical Engineering, City University of Hong Kong（香港城市大学电子工程系） 纠错

Comments 23 pages, 5 figures

Journal ref Proceedings of the 41st International Conference on Machine Learning, 2024

BaryonBridge: Stochastic Interpolant Model for Fast Hydrodynamical Simulations

Benjamin Horowitz, Carolina Cuesta-Lazaro, Omar Yehia

Comments Abstract accepted to the 3rd Machine Learning for Astrophysics Workshop at ICML (July 2025), 3 figures, 6 pages

Improved Exploration in GFlownets via Enhanced Epistemic Neural Networks

Sajan Muhammad, Salem Lahlou

Comments Accepted to the EXAIT Workshop at ICML 2025, and ICoIAS 2025

Adjustment for Confounding using Pre-Trained Representations

Rickmer Schulte, David Rügamer, Thomas Nagler

机构 * Department of Statistics, LMU Munich（统计系，慕尼黑大学） ； Munich Center for Machine Learning (MCML)（慕尼黑机器学习中心） 纠错

Comments Accepted at ICML 2025

Journal ref Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:53557-53580, 2025

Efficient nonparametric statistical inference on population feature importance using Shapley values

Brian D. Williamson, Jean Feng

Comments 15 pages, 3 figures. To be published in the Proceedings of the Thirty-seventh International Conference on Machine Learning (ICML 2020)

Journal ref Proceedings of the 37th International Conference on Machine Learning (2020), PMLR 119:10282-10291

Degeneracy-Aware Pulsar Parameter Estimation from Light Curves via Deep Learning and Test-Time Optimization

Abu Bucker Siddik, Diane Oyen, Soumi De, Greg Olmschenk, Constantinos Kalapotharakos

Comments Accepted at ML4Astro 2025, ICML, Vancouver, CA

Counterfactual Effect Decomposition in Multi-Agent Sequential Decision Making

Stelios Triantafyllou, Aleksa Sukovic, Yasaman Zolfimoselo, Goran Radanovic

机构 * Max Planck Institute for Software Systems（马克斯·普朗克软件系统研究所） 纠错

Comments ICML 2025

Attention Meets Post-hoc Interpretability: A Mathematical Perspective

Gianluigi Lopardo, Frederic Precioso, Damien Garreau

机构 * Universit\'e C\ ote d'Azur, Inria, CNRS 纠错

Comments Accepted at ICML 2024

Journal ref Proceedings of the 41st International Conference on Machine Learning, PMLR 235:32781-32800, 2024

Can Image-To-Video Models Simulate Pedestrian Dynamics?

Aaron Appelle, Jerome P. Lynch

机构 * Duke University（杜克大学） 纠错

Comments Appeared in the ICML 2025 Workshop on Building Physically Plausible World Models, July 2025, https://physical-world-modeling.github.io/

Graph4MM: Weaving Multimodal Learning with Structural Information

Xuying Ning, Dongqi Fu, Tianxin Wei, Wujiang Xu, Jingrui He

机构 * University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校） ； Meta AI ； Rutgers University（罗格斯大学） 纠错

Comments ICML 2025

Progressive Tempering Sampler with Diffusion

Severi Rissanen, RuiKang OuYang, Jiajun He, Wenlin Chen, Markus Heinonen, Arno Solin, José Miguel Hernández-Lobato

机构 * Department of Computer Science, Aalto University, Finland（阿姆斯特丹大学计算机科学系，芬兰） ； Department of Engineering, University of Cambridge, United Kingdom（剑桥大学工程系，英国） ； Department of Empirical Inference, Max Planck Institute for Intelligent Systems, Tübingen, Germany（智能系统研究所经验推断系，德国） 纠错

Comments Accepted for publication at ICML 2025

General agents contain world models

Jonathan Richens, David Abel, Alexis Bellot, Tom Everitt

机构 * Google DeepMind（谷歌DeepMind） 纠错

Comments Accepted ICML 2025. Typos corrected

Gradient Boosting Reinforcement Learning

Benjamin Fuhrer, Chen Tessler, Gal Dalal

机构 * NVIDIA, Tel-Aviv, Israel（NVIDIA，以色列特拉维夫） ； NVIDIA Research, Tel-Aviv, Israel（NVIDIA研究） 纠错

Comments to be published in the Forty-Second International Conference on Machine Learning

Journal ref Proceedings of the 42nd International Conference on Machine Learning ICML 2025, PMLR 267, 17960-17985, Vancouver Canada, 13-19 July 2025

Target Networks and Over-parameterization Stabilize Off-policy Bootstrapping with Function Approximation

Fengdi Che, Chenjun Xiao, Jincheng Mei, Bo Dai, Ramki Gummadi, Oscar A Ramirez, Christopher K Harris, A. Rupam Mahmood, Dale Schuurmans

机构 * Department of Computing Science, University of Alberta（谷歌DeepMind） ； School of Data Science, The Chinese University of Hong Kong, Shenzhen ； School of Computational Science ； Google DeepMind ； CIFAR AI Chair, Amii ； The work was done while the author was at Google. 纠错

Journal ref Proceedings of the 41 st International Conference on Machine Learning, 2024

Kernel-Based Evaluation of Conditional Biological Sequence Models

Pierre Glaser, Steffanie Paul, Alissa M. Hummer, Charlotte M. Deane, Debora S. Marks, Alan N. Amin

机构 * Systems Biology, Harvard Medical School, Boston, USA（系统生物学，哈佛医学院，美国波士顿） ； Harvard Medical School, Broad Institute, Boston, USA（哈佛医学院，Broad研究所，美国波士顿） ； Department of Statistics, University of Oxford, Oxford, UK（统计学系，牛津大学，英国牛津） ； Courant Institute, New York University, New York, USA（Courant研究所，纽约大学，美国纽约） 纠错

Comments 29 pages

Journal ref International Conference on Machine Learning (ICML) 2024

End-to-End Learning Framework for Solving Non-Markovian Optimal Control

Xiaole Zhang, Peiyu Zhang, Xiongye Xiao, Shixuan Li, Vasileios Tzoumas, Vijay Gupta, Paul Bogdan

机构 * Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California, Los Angeles, CA 90089, USA（美国南加州大学明希斯电气与计算机工程系） ； Department of Aerospace Engineering, University of Michigan, Ann Arbor, MI 48109, USA（美国密歇根大学航空航天工程系） ； Elmore Family School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN 47907, USA（美国普渡大学埃尔莫家庭电气与计算机工程学院） 纠错

Journal ref International Conference on Machine Learning (ICML) 2025

Predicting the Subhalo Mass Functions in Simulations from Galaxy Images

Andreas Filipp, Tri Nguyen, Laurence Perreault-Levasseur, Jonah Rose, Chris Lovell, Nicolas Payot, Francisco Villaescusa-Navarro, Yashar Hezaveh

Comments Published as a workshop paper at the ML4Astro Workshop at ICML 2025

Symmetry-Aware GFlowNets

Hohyun Kim, Seunggeun Lee, Min-hwan Oh

机构 * Graduate School of Data Science, Seoul National University, Seoul, Republic of Korea（数据科学研究生院，首尔国立大学，首尔，大韩民国） 纠错

Comments 29 pages; Accepted at ICML 2025

Restoring Calibration for Aligned Large Language Models: A Calibration-Aware Fine-Tuning Approach

Jiancong Xiao, Bojian Hou, Zhanliang Wang, Ruochen Jin, Qi Long, Weijie J. Su, Li Shen

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

Journal ref ICML 2025

Compelling ReLU Networks to Exhibit Exponentially Many Linear Regions at Initialization and During Training

Max Milkert, David Hyde, Forrest Laine

机构 * Department of Computer Science, Vanderbilt University, Nashville TN, USA（范德比尔特大学计算机科学系） ； National Renewable Energy Laboratory, Golden CO, USA（国家可再生能源实验室） 纠错

Comments 24 pages, 17 figures, Poster at ICML 2025

Journal ref Proceedings of the 42nd International Conference on Machine Learning, PMLR Vol. 267, pp. 44175-44198 (ICML 2025)

To Steer or Not to Steer? Mechanistic Error Reduction with Abstention for Language Models

Anna Hedström, Salim I. Amoukou, Tom Bewley, Saumitra Mishra, Manuela Veloso

机构 * J.P. Morgan AI Research（摩根大通AI研究） 纠错

Comments ICML 2025, 22 pages, 16 figures, 5 tables

Journal ref International Machine Learning Conference (ICML) 2025

Of Mice and Machines: A Comparison of Learning Between Real World Mice and RL Agents

Shuo Han, German Espinosa, Junda Huang, Daniel A. Dombeck, Malcolm A. MacIver, Bradly C. Stadie

机构 * Department of Statistics and Data Science, Northwestern University, Evanston, USA（统计学与数据科学系，西北大学，埃文斯顿，美国） ； Department of Mechanical Engineering, Northwestern University, Evanston, USA（机械工程系，西北大学，埃文斯顿，美国） ； Department of Neurology, Northwestern University, Evanston, USA（神经病学系，西北大学，埃文斯顿，美国） 纠错

Comments 19 pages, ICML 2025

Playmate: Flexible Control of Portrait Animation via 3D-Implicit Space Guided Diffusion

Xingpei Ma, Jiaran Cai, Yuansheng Guan, Shenneng Huang, Qiang Zhang, Shunsi Zhang

Journal ref Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:41791-41806, 2025

Position: The Artificial Intelligence and Machine Learning Community Should Adopt a More Transparent and Regulated Peer Review Process

Jing Yang

机构 * University of Southern California（南加州大学） 纠错

Comments ICML 2025; https://papercopilot.com/

Towards Cross-Modal Error Detection with Tables and Images

Olga Ovcharenko, Sebastian Schelter

机构 * Department of XXX, University of YYY, Location, Country（XXX系，YYY大学，地点，国家） ； School of ZZZ, Institute of WWW, Location, Country（ZZZ学院，WWW研究所，地点，国家） 纠错

Journal ref DataWorld Workshop at ICML 2025