ICML

Identifying Metric Structures of Deep Latent Variable Models

Stas Syrota, Yevgen Zainchkovskyy, Johnny Xi, Benjamin Bloem-Reddy, Søren Hauberg

Journal ref Forty-second International Conference on Machine Learning. ICML 2025. Vancouver, Canada. July 13-19, 2025

Flow-of-Options: Diversified and Improved LLM Reasoning by Thinking Through Options

Lakshmi Nair, Ian Trase, Mark Kim

Comments ICML 2025

Boost-and-Skip: A Simple Guidance-Free Diffusion for Minority Generation

Soobin Um, Beomsu Kim, Jong Chul Ye

Comments ICML 2025, 29 pages, 11 figures

Otter: Generating Tests from Issues to Validate SWE Patches

Toufique Ahmed, Jatin Ganhotra, Rangeet Pan, Avraham Shinnar, Saurabh Sinha, Martin Hirzel

Comments Accepted to the main technical track of the International Conference on Machine Learning (ICML), 2025

Safety Reasoning with Guidelines

Haoyu Wang, Zeyu Qin, Li Shen, Xueqian Wang, Dacheng Tao, Minhao Cheng

Comments ICML 2025 paper. The first two authors contributed equally

KBQA-o1: Agentic Knowledge Base Question Answering with Monte Carlo Tree Search

Haoran Luo, Haihong E, Yikai Guo, Qika Lin, Xiaobao Wu, Xinyu Mu, Wenhao Liu, Meina Song, Yifan Zhu, Luu Anh Tuan

Comments Accepted by ICML 2025 main conference

Journal ref ICML 2025

Autonomy-of-Experts Models

Ang Lv, Ruobing Xie, Yining Qian, Songhao Wu, Xingwu Sun, Zhanhui Kang, Di Wang, Rui Yan

Comments Accepted by ICML 2025

AlphaPO: Reward Shape Matters for LLM Alignment

Aman Gupta, Shao Tang, Qingquan Song, Sirou Zhu, Jiwoo Hong, Ankan Saha, Viral Gupta, Noah Lee, Eunki Kim, Siyu Zhu, Parag Agrawal, Natesh Pillai, S. Sathiya Keerthi

机构 * LinkedIn Corporation（LinkedIn公司） ； KAIST AI（KAIST人工智能） ； KAIST（韩国科学技术院） 纠错

Comments 26 pages, 16 figures. Accepted to ICML 2025

DCTdiff: Intriguing Properties of Image Generative Modeling in the DCT Space

Mang Ning, Mingxiao Li, Jianlin Su, Haozhe Jia, Lanmiao Liu, Martin Beneš, Wenshuo Chen, Albert Ali Salah, Itir Onal Ertugrul

Comments ICML 2025

Star Attention: Efficient LLM Inference over Long Sequences

Shantanu Acharya, Fei Jia, Boris Ginsburg

Comments Accepted at ICML 2025

A Generalization Result for Convergence in Learning-to-Optimize

Michael Sucker, Peter Ochs

机构 * University of Tübingen（图宾根大学） ； Saarland University（萨尔兰州立大学） 纠错

Comments spotlight at ICML 2025

Identifying biological perturbation targets through causal differential networks

Menghua Wu, Umesh Padia, Sean H. Murphy, Regina Barzilay, Tommi Jaakkola

Journal ref Proceedings of the 42nd International Conference on Machine Learning, Vancouver, Canada. PMLR 267, 2025

Scaling Large Motion Models with Million-Level Human Motions

Ye Wang, Sipeng Zheng, Bin Cao, Qianshan Wei, Weishuai Zeng, Qin Jin, Zongqing Lu

Comments ICML 2025

On the Vulnerability of Applying Retrieval-Augmented Generation within Knowledge-Intensive Application Domains

Xun Xian, Ganghua Wang, Xuan Bi, Jayanth Srinivasa, Ashish Kundu, Charles Fleming, Mingyi Hong, Jie Ding

机构 * Department of ECE University of Minnesota（电子工程系明尼苏达大学） ； Data Science Institute University of Chicago（数据科学研究所芝加哥大学） ； Carlson School of Management University of Minnesota（卡尔森管理学院明尼苏达大学） ； Department of Surgery University of Minnesota（外科系明尼苏达大学） ； Cisco Research（思科研究） ； School of Statistics University of Minnesota（统计学学院明尼苏达大学） 纠错

Comments Accepted by ICML 2025

Planning in a recurrent neural network that plays Sokoban

Mohammad Taufeeque, Philip Quirke, Maximilian Li, Chris Cundy, Aaron David Tucker, Adam Gleave, Adrià Garriga-Alonso

机构 * Berkeley（伯克利） ； Jane Street 纠错

Comments Mechanistic Interpretability workshop, ICML 2024

Learning Distances from Data with Normalizing Flows and Score Matching

Peter Sorrenson, Daniel Behrend-Uriarte, Christoph Schnörr, Ullrich Köthe

Comments ICML 2025

Binary Hypothesis Testing for Softmax Models and Leverage Score Models

Yuzhou Gu, Zhao Song, Junze Yin

Comments ICML 2025

Primal-Dual Neural Algorithmic Reasoning

Yu He, Ellen Vitercik

Comments The 42nd International Conference on Machine Learning, 2025

NeuronTune: Towards Self-Guided Spurious Bias Mitigation

Guangtao Zheng, Wenqian Ye, Aidong Zhang

Comments Accepted to ICML 2025

Is Your Model Fairly Certain? Uncertainty-Aware Fairness Evaluation for LLMs

Yinong Oliver Wang, Nivedha Sivakumar, Falaah Arif Khan, Rin Metcalf Susa, Adam Golinski, Natalie Mackraz, Barry-John Theobald, Luca Zappella, Nicholas Apostoloff

Comments 9 pages, 8 figures, and 1 table in main paper. Supplementary appendix attached. Accepted at ICML 2025

Confidential Guardian: Cryptographically Prohibiting the Abuse of Model Abstention

Stephan Rabanser, Ali Shahin Shamsabadi, Olive Franzese, Xiao Wang, Adrian Weller, Nicolas Papernot

Comments Proceedings of the 42nd International Conference on Machine Learning

Position: The Future of Bayesian Prediction Is Prior-Fitted

Samuel Müller, Arik Reuter, Noah Hollmann, David Rügamer, Frank Hutter

Comments Accepted as position paper at ICML 2025

Aligning LLMs by Predicting Preferences from User Writing Samples

Stéphane Aroca-Ouellette, Natalie Mackraz, Barry-John Theobald, Katherine Metcalf

Comments Accepted to ICML 2025. 32 pages total: 9 main, 2 references, 21 appendix. arXiv admin note: substantial text overlap with arXiv:2410.06273

X-Transfer Attacks: Towards Super Transferable Adversarial Attacks on CLIP

Hanxun Huang, Sarah Erfani, Yige Li, Xingjun Ma, James Bailey

Comments ICML 2025

Nearly Optimal Sample Complexity for Learning with Label Proportions

Robert Busa-Fekete, Travis Dick, Claudio Gentile, Haim Kaplan, Tomer Koren, Uri Stemmer

机构 * Google Research, New York, USA（谷歌研究, 新 York, 美国） ； Tel Aviv University, Israel（特拉维夫大学, 以色列） 纠错

Comments To appear in ICML 2025

Fixing the Double Penalty in Data-Driven Weather Forecasting Through a Modified Spherical Harmonic Loss Function

Christopher Subich, Syed Zahid Husain, Leo Separovic, Jing Yang

Comments Accepted at ICML 2025

Targeted Unlearning with Single Layer Unlearning Gradient

Zikui Cai, Yaoteng Tan, M. Salman Asif

Comments Accepted to ICML 2025

Model Immunization from a Condition Number Perspective

Amber Yijia Zheng, Cedar Site Bai, Brian Bullins, Raymond A. Yeh

Comments ICML 2025

Understanding Mode Connectivity via Parameter Space Symmetry

Bo Zhao, Nima Dehmamy, Robin Walters, Rose Yu

机构 * Bo Zhao ； Nima Dehmamy ； Robin Walters ； Rose Yu 纠错

Comments 20 pages, 4 figures, ICML 2025

Merge-Friendly Post-Training Quantization for Multi-Target Domain Adaptation

Juncheol Shin, Minsang Seok, Seonggon Kim, Eunhyeok Park

Comments ICML 2025. Code: https://github.com/ewsn1593/HDRQ