Princeton University at ICML 2026

Princeton University researchers will present 12 papers at the 43rd International Conference on Machine Learning (ICML 2026), held July 6-11 in Seoul, South Korea, including four oral presentations and eight spotlight posters. Notable work includes a study showing large language models develop novel social biases through adaptive exploration, and a position paper arguing that benchmark-driven AI cannot foresee certain futures.

The Forty-Third International Conference on Machine Learning https://icml.cc/ , or ICML 2026, will be held July 6-11 in Seoul, South Korea. Below is a roundup of work from Princeton researchers that will be showcased at the conference. Oral Presentation: Large Language Models Develop Novel Social Biases Through Adaptive Exploration https://icml.cc/virtual/2026/oral/71093 Authors: Addison J. Wu , Ryan Liu , Xuechunzi Bai, Thomas L. Griffiths Links: Paper https://arxiv.org/pdf/2511.06148 Oral Presentation: Motion Attribution for Video Generation https://icml.cc/virtual/2026/oral/71049 Authors : Xindi Wu , Despoina Paschalidou, Jun Gao, Antonio Torralba, Laura Leal-Taixé, Olga Russakovsky , Sanja Fidler, Jonathan Lorraine Links : Paper https://arxiv.org/pdf/2601.08828 , Project Website https://research.nvidia.com/labs/sil/projects/MOTIVE/ Oral Presentation: On Computation and Reinforcement Learning https://icml.cc/virtual/2026/oral/71148 Authors : Raj Ghugare , Michał Bortkiewicz , Alicja Ziarko , Benjamin Eysenbach Links : Paper None Oral Presentation: Position: There are futures that benchmark-driven AI cannot see https://icml.cc/virtual/2026/oral/71151 Authors : Sobhan Lotfi, Ava Iranmanesh, Lachin Naghashyar, Ali Shirali, Fateme Haredasht, Sanmi Koyejo, Phil Torr, Yong Suk Lee, Fazl Barez, Joel Lehman, Peter Norvig, Arvind Narayanan Links : Paper https://openreview.net/attachment?id=RzWlUbmRJi&name=pdf Spotlight Poster: AlgoVeri: An Aligned Benchmark for Verified Code Generation on Classical Algorithms https://icml.cc/virtual/2026/poster/61803 Authors : Haoyu Zhao , Ziran Yang , Jiawei Li, Deyuan Mike He, Zenan Li, Chi Jin , Venugopal Veeravalli, Aarti Gupta, Sanjeev Arora Links : Paper, https://arxiv.org/pdf/2602.09464 Project Page https://github.com/haoyuzhao123/algoveri Spotlight Poster: Latent Collaboration in Multi-Agent Systems https://icml.cc/virtual/2026/poster/61180 Authors : Jiaru Zou , Xiyuan Yang, Ruizhong Qiu, Gaotang Li, Katherine Tieu, Pan Lu, Ke Shen, Hanghang Tong, Yejin Choi, Jingrui He, James Zou, Mengdi Wang , Ling Yang Links : Paper https://arxiv.org/pdf/2511.20639 , Project Page https://github.com/Gen-Verse/LatentMAS Spotlight Poster: Rapid Poison: Practical Poisoning Attacks Against the Rapid Response Framework https://icml.cc/virtual/2026/poster/62293 Authors : David Huang , Jaewon Chang, Avidan Shah, Prateek Mittal , Chawin Sitawarin Links : Paper https://arxiv.org/pdf/2606.16242 Adaptively Robust Resettable Streaming https://icml.cc/virtual/2026/poster/66247 Authors : Edith Cohen, Elena Gribelyuk , Jelani Nelson, Uri Stemmer Links : Paper https://arxiv.org/pdf/2601.21989 Position: Agent Should Invoke External Tools ONLY When Epistemically Necessary https://icml.cc/virtual/2026/poster/67227 Authors : Hongru Wang, Cheng Qian, Manling Li, Jiahao Qiu, Boyang XUE, Mengdi Wang , Heng Ji, Amos Storkey, Kam-Fai Wong Links : Paper https://arxiv.org/pdf/2506.00886 AutoTool: Dynamic Tool Selection and Integration for Agentic Reasoning https://icml.cc/virtual/2026/poster/65574 Authors : Jiaru Zou , Ling Yang , Yunzhe Qi, Sirui Chen, Mengting Ai, Ke Shen, Jingrui He, Mengdi Wang Links : Paper https://arxiv.org/pdf/2512.13278 , Project Page https://github.com/Gen-Verse/Open-AgentRL Batched First-Order Methods for Parallel LP Solving in MIP https://icml.cc/virtual/2026/poster/65795 Authors : Nicolas Blin, Stefano Gualandi, Christopher Maes, Andrea Lodi, Bartolomeo Stellato Links : Paper https://arxiv.org/pdf/2601.21990 Position: The Case for Theory-Level Autoformalization https://icml.cc/virtual/2026/poster/67212 Authors : Marcus Min, Mike He , Zhaoyu Li, Zixuan Yi, Sharad Malik , Aarti Gupta , Xujie Si, Osbert Bastani Links : Project Page https://marcusm117.github.io/publications/ CodeClash: Benchmarking Goal-Oriented Software Engineering https://icml.cc/virtual/2026/poster/63924 Authors : John Yang, Kilian Lieret , Joyce Yang, Carlos Jimenez , Muhtasham Oblokulov, Aryan Siddiqui, Ofir Press , Ludwig Schmidt, Diyi Yang Links : Paper https://arxiv.org/pdf/2511.00839 Conformal Prediction for Early Stopping in Mixed Integer Optimization https://icml.cc/virtual/2026/poster/61715 Authors : Stefan Clarke , Bartolomeo Stellato Links : Paper https://arxiv.org/pdf/2602.01476 Consistent Zero-Shot Imitation with Contrastive Goal Inference https://icml.cc/virtual/2026/poster/62891 Authors : Kathryn Wantlin , Chongyi Zheng , Benjamin Eysenbach Links : Paper https://arxiv.org/pdf/2510.17059 Coupled Training with Privileged Features and Unlabeled Data Authors : Jiahao Shi , Omar Hagrass , Jason Klusowski Links : Paper None DevEvol: Benchmarking LLM Agents on Continuous Software Evolution https://icml.cc/virtual/2026/poster/65024 Authors : Gangda Deng, Zhaoling Chen, Zhongming Yu, Haoyang Fan, Yuhong Liu, Yuxin Yang, Dhruv Parikh, Rajgopal Kannan, Le Cong, Mengdi Wang , Qian Zhang, Viktor Prasanna, Robert Tang, Xingyao Wang Links : Paper https://openreview.net/attachment?id=HeRPNfJaod&name=originally submitted PDF Position: Digital Agents Require Unified Agent-Native Environments https://icml.cc/virtual/2026/poster/67134 Authors : Yiran Wu, Jiale Liu, Jieyu Zhang, Yaolun Zhang, Shilong Liu , Chi Wang, Mengdi Wang , Huazheng Wang, Qingyun Wu Links : Paper https://openreview.net/pdf?id=XZaRHkbc2u f-Divergence Regularized RLHF: Two Tales of Sampling and Unified Analyses https://icml.cc/virtual/2026/poster/63348 Authors : Di Wu, Chengshuai Shi , Jing Yang, Cong Shen Links : Paper https://arxiv.org/pdf/2605.06977 Finding the Minimal Parameter Budget for Implicit Reasoning: A Data Complexity Driven Scaling Law for Language Models https://icml.cc/virtual/2026/poster/62190 Authors : Xinyi Wang , Shawn Tan, Shenbo Xu, Mingyu Jin, William Wang, Rameswar Panda, Yikang Shen Links : Paper None FLARE-AI: Flaw Reporting for AI https://icml.cc/virtual/2026/poster/65171 Authors : Shayne Longpre, Elaine Zhu, Carson Ezell, Avijit Ghosh, Sean McGregor, Kevin Paeth, Kevin Klyman, Sayash Kapoor , Rishi Bommasani, Ruth Elisabeth Appel, Gregory Strom, Lauren McIlvenny, Mark Jaycox, Peter Slattery, Nathan Butters, Arvind Narayanan , Percy Liang, Alex Pentland Links : Paper https://arxiv.org/html/2606.31567v1 FrontierCS: Evolving Challenges for Evolving Intelligence https://icml.cc/virtual/2026/poster/61419 Authors : Qiuyang Mang, Wenhao Chai , Zhifei Li, Huanzhi Mao, Shang Zhou, Alexander Du, Hanchen Li, Shu Liu, Edwin Chen, Yichuan Wang, Xieting Chu, Zerui Cheng , Yuan Xu, Tian Xia, Zirui Wang, Tianneng Shi, Jianzhu Yao , Yilong Zhao, Qizheng Zhang, Charlie Ruan, Zeyu Shen , Kaiyuan Liu, Zhaoyang Hong, Alex Gu, Ziyi Zhang, Runyuan He, Dong Xing, Zerui Li, Zirong Zeng, Yige Jiang, Lufeng Cheng, Ziyi Zhao, Youran Sun, Suyang Zhong, Junpeng Wang, Donglin Li, Wenyuan Huang, Jialiang Gu, Wesley Zheng, Wangmeiyu Zhang, Ruyi Ji, Xuechang Tu, Zihan Zheng, Zhaozi Wang, Zexing Chen, Jingbang Chen, Jialu Zhang, Aleksandra Korolova , Peter Henderson , Pramod Viswanath , Vijay Ganesh, Saining Xie, Zhuang Liu, Dawn Song, Sewon Min, Ion Stoica, Joseph E Gonzalez, Jingbo Shang, Alvin Cheung Links : Paper https://arxiv.org/pdf/2512.15699 , Project Page https://frontier-cs.org/ GASS: Geometry-Aware Spherical Sampling for Disentangled Diversity Enhancement in Text-to-Image Generation https://icml.cc/virtual/2026/poster/62652 Authors : Ye Zhu , Kaleb Newman , Johannes Lutzeyer, Adriana Romero-Soriano, Michal Drozdzal, Olga Russakovsky Links : Paper https://arxiv.org/pdf/2602.17200 Generalization Bounds for Discrete Diffusion: Statistical Advantage of Masking https://icml.cc/virtual/2026/poster/64332 Authors : Zixuan Zhang , Hengyu Fu, Zhuoran Yang, Mengdi Wang , Tuo Zhao, Minshuo Chen Links : Paper https://openreview.net/pdf?id=Ofhq7nBVHu The Geometry of Representational Failures in Vision Language Models https://icml.cc/virtual/2026/poster/63080 Authors : Daniele Savietto, Declan Campbell , André Panisson, Marco Nurisso, Giovanni Petri, Jonathan Cohen , Alan Perotti Links : Paper https://arxiv.org/pdf/2602.07025 Greedy Coordinate Diffusion: Effective and Semantically Coherent Adversarial Attacks via Diffusion Guidance https://icml.cc/virtual/2026/poster/63755 Authors : Bohdan Turbal , Blossom Metevier , Max Springer , Aleksandra Korolova Links : Paper https://arxiv.org/pdf/2606.15531 Hyperparameter Transfer with Mixture-of-Expert Layers https://icml.cc/virtual/2026/poster/62595 Authors : Tianze Jiang , Blake Bordelon, Cengiz Pehlevan, Boris Hanin Links : Paper https://arxiv.org/pdf/2601.20205 A Hypertoroidal Covering for Perfect Color Equivariance https://icml.cc/virtual/2026/poster/64636 Authors : Yulong Yang , Zhikun Xu , Yaojun Li , Christine Allen-Blanchette Links : Paper https://arxiv.org/pdf/2603.04256 LatentChem: From Textual CoT to Latent Thinking in Chemical Reasoning https://icml.cc/virtual/2026/poster/63674 Authors : Xinwu Ye, Yicheng Mao, Jia Zhang, Yimeng Yoyo Liu, Hao Li, Fang Wu, Zhiwei Li, Yuxuan Liao, Zehong Wang, Zhiyuan Liu, Zhenfei Yin, Li Yuan, Phil Torr, Huan Sun, Xiangxiang Zeng, Mengdi Wang , Le Cong, Shenghua Gao, Robert Tang Links : Paper https://arxiv.org/pdf/2602.07075 , Project Page https://github.com/xinwuye/LatentChem Learning to Perceive the World Through Control: Empowerment-Based Representation Learning https://icml.cc/virtual/2026/poster/63198 Authors : Mahsa Bastankhah , Sophie Broderick , Benjamin Eysenbach Links : Paper https://arxiv.org/pdf/2605.30656 Less Is More: Fast and Accurate Reasoning with Cross-Head Unified Sparse Attention https://icml.cc/virtual/2026/poster/61079 Authors : Lijie Yang , Zhihao Zhang, Arti Jain, Shijie Cao, Baihong Yuan, Yiwei Chen, Zhihao Jia, Ravi Netravali Links : Paper https://arxiv.org/pdf/2508.07101 , Project Page https://github.com/DerrickYLJ/LessIsMore/tree/master Lower Bounds for Frank-Wolfe on Strongly Convex Sets https://icml.cc/virtual/2026/poster/66478 Authors : Jannis Halbey, Daniel Deza , Max Zimmer, Christophe Roux, Bartolomeo Stellato , Sebastian Pokutta Links : Paper https://arxiv.org/pdf/2602.04378 MEMO: Memory-Augmented Model Context Optimization for Robust Multi-Turn Multi-Agent LLM Games https://icml.cc/virtual/2026/poster/62950 Authors : Yunfei Xie, Kevin Wang, Bobby Cheng, Jianzhu Yao , Zhizhou Sha, Alexander Duffy, Yihan Xi, Hongyuan Mei, Cheston Tan, Chen Wei, Pramod Viswanath, Zhangyang “Atlas” Wang Links : Paper https://arxiv.org/pdf/2603.09022 On Path to Multimodal Historical Reasoning: HistBench and HistAgent https://icml.cc/virtual/2026/poster/62318 Authors : Jiahao Qiu , Fulian Xiao, Yimin Wang, Yuchen Mao, Yijia Chen, Xinzhe Juan, Siran Wang, Xuan Qi, Tongcheng Zhang, Zixin Yao, Jiacheng Guo , Yifu Lu , Charles Argon, Jundi Cui, Daixin Chen, Junran Zhou, Shuyao Zhou, Zhanpeng Zhou, Ling Yang , Shilong Liu , Hongru Wang, Kaixuan Huang , Xun Jiang, Xi Gao, Mengdi Wang Links : Paper https://arxiv.org/pdf/2505.20246 Opportunistic Expert Activation: Batch-Aware Expert Routing for Faster Decode Without Retraining https://icml.cc/virtual/2026/poster/62958 Authors : Costin-Andrei Oncescu, Qingyang Wu, Wai Tong Chung, Tsai-chuan Wu, Bryan Gopal, Junxiong Wang, Tri Dao , Ben Athiwaratkun Links : Paper https://arxiv.org/pdf/2511.02237 Position: Multiplicity is an Inevitable and Inherent Challenge in Multimodal Learning https://icml.cc/virtual/2026/poster/67073 Authors : Sanghyuk Chun , Olga Russakovsky Links : Paper https://arxiv.org/pdf/2505.19614 The Power of Power Law: Asymmetry Enables Compositional Reasoning https://icml.cc/virtual/2026/poster/64765 Authors : Zixuan Wang , Xingyu Dang, Jason Lee, Kaifeng Lyu Links : Paper https://arxiv.org/pdf/2604.22951 Prioritize the Process, Not Just the Outcome: Rewarding Latent Thought Trajectories Improves Reasoning in Looped Language Models https://icml.cc/virtual/2026/poster/62869 Authors : Jonathan Williams, Esin Tureci, Olga Russakovsky Links : Paper https://arxiv.org/html/2602.10520v1 , Project Page https://github.com/jonwill8/RLTT Retaining by Doing: The Role of On-Policy Data in Mitigating Forgetting https://icml.cc/virtual/2026/poster/64375 Authors : Howard Chen , Noam Razin , Karthik Narasimhan , Danqi Chen Links : Paper https://arxiv.org/pdf/2510.18874 Rethinking Thinking Tokens: LLMs as Improvement Operators Authors : Lovish Madaan, Aniket Didolkar, Suchin Gururangan, John Quan, Ruan Silva, Russ Salakhutdinov, Manzil Zaheer, Sanjeev Arora , Anirudh Goyal Links : Paper https://arxiv.org/pdf/2510.01123 RLAnything: Forge Environment, Policy, and Reward Model in Completely Dynamic RL System https://icml.cc/virtual/2026/poster/64832 Authors : Yinjie Wang , Tianbao Xie , Ke Shen , Mengdi Wang , Ling Yang Links : Paper https://arxiv.org/pdf/2602.02488 , Project Page https://github.com/Gen-Verse/Open-AgentRL SMILE: Extended Deep Submodular Function-Based Instruction and In-context Learning Demonstration Selection https://icml.cc/virtual/2026/poster/64076 Authors : Zihan Chen, Chengshuai Shi , Song Wang, Jundong Li, Cong Shen Links : Paper https://openreview.net/pdf?id=R8hivF3AkV Statistical-Computational Trade-offs for Recursive Adaptive Partitioning Estimators https://icml.cc/virtual/2026/poster/71801 Authors : Yan Shuo Tan, Jason Klusowski , Krishnakumar Balasubramanian Links : Paper https://arxiv.org/pdf/2411.04394 SWE-fficiency: Can Language Models Optimize Real-World Repositories on Real Workloads? https://icml.cc/virtual/2026/poster/66745 Authors : Jeffrey Ma, Milad Hashemi, Amir Yazdanbakhsh, Kevin Swersky, Ofir Press , Enhui Li, Vijay Janapa Reddi, Parthasarathy Ranganathan Links : Paper https://arxiv.org/pdf/2511.06090 , Project Page https://swefficiency.com/ τ-Knowledge: Evaluating Conversational Agents over Unstructured Knowledge https://icml.cc/virtual/2026/poster/63427 Authors : Quan Shi, Alexandra Zytek, Pedram Razavi, Karthik Narasimhan , Victor Barres Links : Paper https://arxiv.org/pdf/2603.04370 , Project Page https://sierra.ai/blog/tau-knowledge t-Voice: Benchmarking Full-Duplex Voice Agents on Real-World Domains https://icml.cc/virtual/2026/poster/66590 Authors : Soham Ray, Keshav Dhandhania, Victor Barres, Karthik Narasimhan Links : Paper https://arxiv.org/pdf/2603.13686 Temporal Context Reinstatement Drives Episodic-Like Order Memory in Long-Context Language Models https://icml.cc/virtual/2026/poster/61179 Authors : Mathis Pink, Vy Vo, Qinyuan Wu, Jianing Mu, Javier Turek, Uri Hasson , Kenneth Norman , Sebastian Michelmann, Alexander Huth, Mariya Toneva Topology-Preserving Neural Operator Learning via Hodge Decomposition https://icml.cc/virtual/2026/poster/61237 Authors : Dongzhe Zheng , Tao Zhong , Christine Allen-Blanchette Links : Paper https://arxiv.org/pdf/2605.13834 , Project Page https://github.com/ContinuumCoder/Hodge-Spectral-Duality Towards a Science of AI Agent Reliability https://icml.cc/virtual/2026/poster/66364 Authors : Stephan Rabanser , Sayash Kapoor , Peter Kirgis , Kangheng Liu , Saiteja Utpala , Arvind Narayanan Links : Paper https://arxiv.org/pdf/2602.16666 Training LLM Agents to Empower Humans https://icml.cc/virtual/2026/poster/61627 Authors : Evan Ellis, Vivek Myers, Jens Tuyls, Sergey Levine, Anca Dragan, Benjamin Eysenbach Links : Paper https://arxiv.org/pdf/2510.13709 TritonGym: A Benchmark for Agentic LLM Workflows in Triton GPU Code Generation https://icml.cc/virtual/2026/poster/65003 Authors : Yue Guan, Yichen Lin, Xu Zhao, Jianzhu Yao, Xinwei Qiang, Zhongkai Yu, Pramod Viswanath , Yufei Ding, Adnan Aziz Links : Paper https://openreview.net/pdf?id=oaKd1fVgWc , Project Page https://tritongym.picasso-lab.com/ Universal Learning of Nonlinear Dynamics https://icml.cc/virtual/2026/poster/65255 Authors : Evan Dogariu, Anand Brahmbhatt , Elad Hazan Links : Paper https://arxiv.org/pdf/2508.11990