Center for Advanced Intelligence Project

43 papers have been accepted at the International Conference on Machine Learning (ICML) 2026, a major conference on Artificial Intelligence (July 6-11, 2026, Seoul, South Korea).

For more details, please refer to the link below.
[Website] https://icml.cc/Conferences/2026

[Acceptance rate]
There are 6,352 accepted papers from 23,918 submissions, leading to a 26.6% percent acceptance rate.
*Spotlight Posters:2.2% (536/23,918)

Spotlight

• Fair Classification with Efficient and Post-hoc Controllable Fairness-Accuracy Trade-off
  Maaya Sakata (University of Tsukuba / RIKEN AIP)
  Kazuto Fukuchi (University of Tsukuba / RIKEN AIP)
• SVRG and Beyond via Posterior Correction
  Nico Daheim (Technical University of Darmstadt / hessian.AI)
  Thomas Möllenhoff (RIKEN AIP)
  Ming Liang Ang (University College London)
  Mohammad Emtiyaz Khan (RIKEN AIP)

Regular

• A Decision-Theoretic View of Test-Time Training: When, How Far, and Which Directions to Adapt
  Tomoya Wakayama (RIKEN AIP)
• A Perturbation Approach to Unconstrained Linear Bandits
  Andrew Jacobsen (Università degli Studi di Milano / Politecnico di Milano)
  Dorian Baudry (Inria, Univ. Grenoble Alpes, Grenoble INP, CNRS, LIG)
  Shinji Ito (University of Tokyo / RIKEN AIP)
  Nicolò Cesa-Bianchi (Università degli Studi di Milano / Politecnico di Milano)
• Approximation Theory for Lipschitz Continuous Transformers
  Takashi Furuya (Doshisha University, RIKEN AIP)
  Davide Murari (University of Cambridge)
  Carola-Bibiane Schönlieb (University of Cambridge)
• Attention with Routed-Memory for Learnable Sparse Control
  Qiuhao Zeng (University of Toronto)
  Jerry Huang (University of Montreal)
  Peng Lu (University of Montreal)
  Ruiyi Fang (University of Western Ontario)
  Gezheng Xu (University of Western Ontario)
  Zihao Jing (University of Western Ontario)
  Yufei Cui (McGill University)
  Charles Ling (University of Western Ontario)
  Gang Niu (RIKEN AIP)
  Boyu Wang (University of Western Ontario)
• Bayesian Tensor Decomposition with Diffusion Model Prior
  Zerui Tao (RIKEN AIP)
  Qibin Zhao (RIKEN AIP)
• Calibrating Uncertainty for Zero-Shot Adversarial CLIP
  Wenjing Lu (Shanghai Jiao Tong University / RIKEN AIP)
  Zerui Tao (RIKEN AIP)
  Yuning Qiu (RIKEN AIP)
  Dongping Zhang (Guangdong University of Technology / RIKEN AIP)
  Yang Yang (Shanghai Jiao Tong University)
  Qibin Zhao (RIKEN AIP)
• Compositional Behavioral Semantics and Metrics for State Abstraction in Reinforcement Learning
  Yivan Zhang (University of Tokyo / RIKEN AIP)
  Ziyan Luo (Mila / McGill University)
  Manuel Baltieri (Araya Inc. / University of Sussex)
• Data- and Variance-dependent Regret Bounds for Online Tabular MDPs
  Mingyi Li (University of Tokyo)
  Taira Tsuchiya (University of Tokyo / RIKEN AIP)
  Kenji Yamanishi (University of Tokyo)
• Decomposing the Basic Abilities of Large Language Models: Mitigating Cross-Task Interference in Multi-Task Instruct-Tuning
  Bing Wang (Jilin University)
  Ximing Li (Jilin University / RIKEN AIP)
  Changchun Li (Jilin University)
  Jinjin Chi (Jilin University)
  Gang Niu (RIKEN AIP)
  Masashi Sugiyama (RIKEN AIP / University of Tokyo)
• Dichotomy of Feature Learning and Unlearning: Fast-Slow Analysis on Neural Networks with Stochastic Gradient Descent
  Shota Imai (University of Tokyo)
  Sota Nishiyama (University of Tokyo)
  Masaaki Imaizumi (RIKEN AIP / University of Tokyo)
• Distortion of AI Alignment Revisited: RLHF Is a Decent Utilitarian Aligner.
  Kazusato Oko (UC Berkeley / RIKEN AIP)
  Annie S Ulichney (UC Berkeley)
  Nika Haghtalab (UC Berkeley)
  Han Bao (The Institute of Statistical Mathematics / Tohoku University / RIKEN AIP)
• EEG-Based Multimodal Learning via Hyperbolic Mixture-of-Curvature Experts
  Runhe Zhou* (Nanyang Technological University)
  Shanglin Li* (ATR / RIKEN AIP)
  Guanxiang Huang (University of Cambridge)
  Xinliang Zhou (Nanyang Technological University)
  Qibin Zhao (RIKEN AIP)
  Motoaki Kawanabe (ATR)
  Yi Ding (Nanyang Technological University)
  Cuntai Guan (Nanyang Technological University)
  * Equal contribution
• Finite and Corruption-Robust Regret Bounds in Online Inverse Linear Optimization under M-Convex Action Sets.
  Taihei Oki (Hokkaido University / RIKEN AIP)
  Shinsaku Sakaue (CyberAgent / National Institute of Informatics / RIKEN AIP)
• Gradient Regularization Prevents Reward Hacking in Reinforcement Learning from Human Feedback and Verifiable Rewards
  Johannes Ackermann (University of Tokyo / RIKEN AIP)
  Michael Noukhovitch (University of Montreal)
  Takashi Ishida (RIKEN AIP / University of Tokyo)
  Masashi Sugiyama (RIKEN AIP / University of Tokyo)
• High-Dimensional Learning Dynamics of Quantized Models with Straight-Through Estimator
  Yuma Ichikawa (Fujitsu Limited / RIKEN AIP)
  Shuhei Kashiwamura (University of Tokyo / NTT Research)
  Ayaka Sakata (Ochanomizu University / RIKEN AIP)
• How Can I Publish My LLM Benchmark Without Giving the True Answers Away?
  Takashi Ishida (RIKEN AIP / University of Tokyo)
  Thanawat Lodkaew (University of Tokyo / RIKEN AIP)
  Ikko Yamane (ENSAI / CREST)
• In-Context Learning Is Provably Bayesian Inference: A Generalization Theory for Meta-Learning
  Tomoya Wakayama (RIKEN AIP)
  Taiji Suzuki (University of Tokyo / RIKEN AIP)
• Incorporating Importance Weighting in Optimal Transport Based Domain Alignment
  Okan Koç (RIKEN AIP)
  Alexander Soen (KTH Royal Institute of Technology / Australian National University)
  Shanglin Li (Technische Universität Berlin / ATR / RIKEN AIP)
  Masashi Sugiyama (RIKEN AIP / University of Tokyo)
• Inference-Aware Meta-Alignment of LLMs via Non-Linear GRPO Authors list:
  Shokichi Takakura (LY Corporation)
  Akifumi Wachi (LY Corporation)
  Rei Higuchi (University of Tokyo / RIKEN AIP)
  Kohei Miyaguchi (LY Corporation)
  Taiji Suzuki (University of Tokyo / RIKEN AIP)
• Inference-time Alignment with Rewards in Besov Spaces: Provable Advantages of Feature Learning and Multi-Step Policy Updates Authors list:
  Naoki Nishikawa (University of Tokyo / RIKEN AIP)
  Taiji Suzuki (University of Tokyo / RIKEN AIP)
• Information-Theoretic Generalization Bounds for VAEs: A Role of Encoder and Latent
  Variable
  Futoshi Futami (University of Osaka / RIKEN AIP / University of Tokyo)
  Masahiro Fujisawa (University of Osaka / RIKEN AIP)
• Joint Model and Data Sparsification via the Marginal Likelihood
  Alexander Timans (University of Amsterdam)
  Thomas Möllenhoff (RIKEN AIP)
  Christian Andersson Naesseth (University of Amsterdam)
  Mohammad Emtiyaz Khan (RIKEN AIP)
  Eric Nalisnick (Johns Hopkins University)
• Last-Iterate Convergence of Regularized Gradient Methods for Stochastic Monotone Variational Inequalities
  Shinji Ito (University of Tokyo / RIKEN AIP)
  Taira Tsuchiya (University of Tokyo / RIKEN AIP)
  Kaito Ariu (CyberAgent)
  Kenshi Abe (CyberAgent)
• Log-Normal Multiplicative Dynamics for Stable Low-Precision Training of Large Networks
  Keigo Nishida (RIKEN BDR / RIKEN AIP)
  Eren Mehmet Kıral (RIKEN AIP / Keio University)
  Kenichi Bannai (Keio University / RIKEN AIP)
  Mohammad Emtiyaz Khan (RIKEN AIP)
  Thomas Möllenhoff (RIKEN AIP)
• Mesh Field Theory: Port–Hamiltonian Formulation of Mesh-Based Physics
  Satoshi Noguchi (JAMSTEC / RIKEN AIP)
  Yoshinobu Kawahara (University of Osaka / RIKEN AIP)
• MTNL: A Unified Modeling Perspective for Enhancing Tensor Network Learning
  Junhua Zeng (Inner Mongolia University)
  Yuning Qiu (RIKEN AIP)
  Binghua Li (Juntendo University)
  Chao Li (RIKEN AIP)
  Qibin Zhao (RIKEN AIP)
  Guoxu Zhou (Guangdong University of Technology)
• Non-Stationary Online Structured Prediction with Surrogate Losses.
  Shinsaku Sakaue (CyberAgent / National Institute of Informatics / RIKEN AIP)
  Han Bao (The Institute of Statistical Mathematics / Tohoku University / RIKEN AIP)
  Yuzhou Cao (Nanyang Technological University)
• Positive-Unlabeled Learning with Extreme Scarcity of Labeled Positives
  Yuanchao Dai (Jilin University)
  Ximing Li (Jilin University / RIKEN AIP)
  Wei Wang (RIKEN AIP)
  Changchun Li (Jilin University)
  Gang Niu (RIKEN AIP)
  Masashi Sugiyama (RIKEN AIP / University of Tokyo)
• Positive–Unlabeled Reinforcement Learning Distillation for On-Premise Small Models
  Zhiqiang Kou (Hong Kong Polytechnic University)
  Junyang Chen (Southeast University)
  Xin-Qiang Cai (RIKEN AIP)
  Xiaobo Xia (National University of Singapore)
  Ming-Kun Xie (RIKEN AIP)
  Dong-Dong Wu (University of Tokyo / RIKEN AIP)
  Biao Liu (Southeast University)
  Yuheng Jia (Southeast University)
  Xin Geng (Southeast University)
  Masashi Sugiyama (RIKEN AIP / University of Tokyo)
  Tat-Seng Chua (National University of Singapore)
• Proteo-R1: Thinking Foundation Models for De Novo Protein Binder Design
  Fang Wu (Stanford University)
  Li Erran Li (AWS AI / Columbia University)
  Weihao Xuan (University of Tokyo / RIKEN AIP)
  Heli Qi (RIKEN AIP / Waseda University)
  Zeqi Zhou (Brown University)
  Hanqun Cao (Chinese University of Hong Kong)
  Heng-Jui Chang (Stanford University)
  Haokai Zhao (University of New South Wales)
  Jian Ma (Shanghai Jiao Tong University)
  Zijian Carl Ma (Stanford University)
  Yu-Chi Cheng (Harvard University)
  Xiangru Tang (Yale University)
  Zehong Wang (University of Notre Dame)
  Kuan Pang (Stanford University)
  Hanchen Wang (Stanford University)
  Kejun Ying (Stanford University)
  Pan Lu (Stanford University)
  Chiho Im (Stanford University)
  Seungju Han (Stanford University)
  Peng Xia (University of North Carolina at Chapel Hill)
  Yinxi Li (Waterloo University)
  Guanlue Li (University of Hamburg)
  Tinson Xu (University of Chicago)
  Deyao Zhu (ByteDance)
  Pheng-Ann Heng (Chinese University of Hong Kong)
  Naoto Yokoya (University of Tokyo / RIKEN AIP)
  Masashi Sugiyama (University of Tokyo / RIKEN AIP)
  Jure Leskovec (Stanford University)
  Yejin Choi (Stanford University)
• Provable Sample Efficiency of Curriculum Post-Training for Transformer Reasoning
  Dake Bu (City University of Hong Kong / RIKEN AIP / A*STAR)
  Wei Huang (RIKEN AIP / The Institute of Statistical Mathematics)
  Andi Han (University of Sydney)
  Atsushi Nitanda (A*STAR / Nanyang Technological University)
  Hau-San Wong (City University of Hong Kong)
  Qingfu Zhang (City University of Hong Kong)
  Taiji Suzuki (University of Tokyo / RIKEN AIP)
• Refining Dual Spectral Sparsity in Transformed Tensor Singular Values
  Andong Wang (RIKEN AIP)
  Yuning Qiu (RIKEN AIP)
  Haonan Huang (RIKEN AIP)
  Zhong Jin (China University of Petroleum-Beijing at Karamay)
  Guoxu Zhou (Guangdong University of Technology)
  Qibin Zhao (RIKEN AIP)
• Revisiting Regularized Policy Optimization for Stable and Efficient Reinforcement
  Learning in Two-Player Games
  Kazuki Ota (University of Tokyo)
  Takayuki Osa (RIKEN AIP)
  Motoki Omura (University of Tokyo)
  Tatsuya Harada (University of Tokyo / RIKEN AIP)
• Sign Lock-In: Randomly Initialized Weight Signs Persist and Bottleneck Sub-Bit Model Compression
  Akira Sakai (Fujitsu Limited / Tokai University)
  Yuma Ichikawa (Fujitsu Limited / RIKEN AIP)
• Spectral Gradient Descent Mitigates Anisotropy-Driven Misalignment: A Case
  Study in Phase Retrieval
  Guillaume Braun (RIKEN AIP)
  Han Bao (The Institute of Statistical Mathematics)
  Wei Huang (RIKEN AIP)
  Masaaki Imaizumi (RIKEN AIP / University of Tokyo)
• State Space Model with Continuous Limit of HiPPO Matrix: Eigenvalue Analysis and Explicit Solution Formula
  Atsushi Takabatake (Kyushu University / RIKEN AIP),
  Takaharu Yaguchi (Kyushu University / RIKEN AIP)
• Towards Disentangled Preference Optimization Dynamics
  Wei Chen (South China University of Technology / RIKEN AIP)
  Yubing Wu (South China University of Technology)
  Junmei Yang (South China University of Technology)
  Delu Zeng (South China University of Technology)
  Qibin Zhao (RIKEN AIP)
  John Paisley (Columbia University)
  Min Chen (South China University of Technology)
  Zhou Wang (University of Waterloo)
• Two Modalities Are Better Than One: Efficient Adversarial Purification via Multimodal Diffusion Models
  Mingyuan Bai (RIKEN AIP)
  Wei Huang (RIKEN AIP)
  Tenghui Li (Guangdong University of Technology)
  Andong Wang (RIKEN AIP)
  Chao Li (RIKEN AIP)
  Cesar F Caiafa (CONICET / University of Buenos Aires / RIKEN AIP)
  Junbin Gao (University of Sydney)
  Qibin Zhao (RIKEN AIP)
• Why Agentic Theorem Prover Works: A Statistical Provability Theory of Mathematical Reasoning Models
  Sho Sonoda (RIKEN AIP / CyberAgent)
  Shunta Akiyama (CyberAgent)
  Yuya Uezato (CyberAgent)
• Zero Flow Encoders
  Yakun Wang (Bristol)
  Leyang Wang (RIKEN AIP)
  Song Liu (Bristol)
  Taiji Suzuki (University of Tokyo / RIKEN AIP)
• Position: Agentic AI Orchestration Should Be Bayes-Consistent
  Theodore Papamarkou (PolyShape / NTUA)
  Pierre Alquier (ESSEC Business School)
  Matthias Bauer
  Wray Buntine (VinUniversity)
  Andrew Davison (Imperial College London)
  Gintare Karolina Dziugaite (Mila – Quebec AI Institute)
  Maurizio Filippone (KAUST)
  Andrew Y. K. Foong (Mayo Clinic)
  Vincent Fortuin (TU Nuremberg / Helmholtz AI)
  Dimitris Fouskakis (NTUA)
  Eyke Hüllermeier (LMU Munich)
  Theofanis Karaletsos (CZI Science)
  Mohammad Emtiyaz Khan (RIKEN AIP)
  Nikita Kotelevskii (MBZUAI)
  Salem Lahlou (MBZUAI)
  Yingzhen Li (Imperial College London)
  Fang Liu (University of Notre Dame)
  Clare Lyle
  Thomas Möllenhoff (RIKEN AIP)
  Konstantina Palla (Spotify)
  Maxim Panov (MBZUAI)
  Yusuf Sale (LMU Munich)
  Kajetan Schweighofer (Cognizant AI Lab)
  Artem Shelmanov (MBZUAI)
  Siddharth Swaroop (University College London)
  Martin Trapp (KTH Royal Institute of Technology)
  Willem Waegeman (Ghent University)
  Andrew Gordon Wilson (New York University)
  Alexey Zaytsev (BIMSA)

Related Laboratories