Center for Advanced Intelligence Project

44 papers were accepted at NeurIPS 2025, which is known as a top conference on machine learning.

For more details, please refer to the link below:

[NeurIPS 2025]

There were 21,575 full paper submissions to NeurIPS this year, of which the program committee accepted 24.52%* for presentation at the conference.
*Poster 4,525　Spotlight 688　Oral 77

Spotlight

• Any-stepsize Gradient Descent for Separable Data under Fenchel–Young Losses
  Han Bao (Institute of Statistical Mathematics)
  Shinsaku Sakaue (CyberAgent / The University of Tokyo / RIKEN AIP)
  Yuki Takezawa (Kyoto University / OIST)
• From Shortcut to Induction Head: How Data Diversity Shapes Algorithm Selection in Transformers
  Ryotaro Kawata (The University of Tokyo / RIKEN AIP)
  Yujin Song (Exawizards Inc.)
  Alberto Bietti (Flatiron Institute)
  Naoki Nishikawa (The University of Tokyo / RIKEN AIP)
  Taiji Suzuki (The University of Tokyo / RIKEN AIP)
  Samuel Vaiter (CNRS)
  Denny Wu (New York University / Flatiron Institute)
• GeoSVR: Taming Sparse Voxels for Geometrically Accurate Surface Reconstruction
  Jiahe Li (Beihang University)
  Jiawei Zhang (Beihang University)
  Youmin Zhang (Rawmantic AI)
  Xiao Bai (Beihang University)
  Jin Zheng (Beihang University)
  Xiaohan Yu (Macquarie University)
  Lin Gu (RIKEN AIP)
• Variational Learning Finds Flatter Solutions at the Edge of Stability,
  Avrajit Ghosh* (Michigan State University)++
  Bai Cong (RIKEN AIP / Institute of Science Tokyo)
  Rio Yokota (Institute of Science Tokyo)
  Saiprasad Ravishankar (Michigan State University)
  Rongrong Wang (Michigan State University)
  Molei Tao (Georgia Institute of Technology)
  Mohammad Emtiyaz Khan (RIKEN AIP)
  Thomas Möllenhoff (RIKEN AIP)
  *work done during internship at RIKEN AIP

Poster

• A Closer Look to Positive-Unlabeled Learning from Fine-grained Perspectives: An Empirical Study
  Yuanchao Dai (Jilin University)
  Zhengzhang Hou (Jilin University)
  Changchun Li (Jilin University)
  Yuanbo Xu (Jilin University)
  En Wang (Jilin University)
  Ximing Li (Jilin University / RIKEN AIP);
• Adapting to Stochastic and Adversarial Losses in Episodic MDPs with Aggregate Bandit Feedback
  Shinji Ito (The University of Tokyo / RIKEN AIP)
  Kevin Jamieson (University of Washington)
  Haipeng Luo (University of Southern California)
  Arnab Maiti (University of Washington)
  Taira Tsuchiya (The University of Tokyo / RIKEN AIP)
  * Alphabetical order
• A Frustratingly Simple Yet Highly Effective Attack Baseline: Over 90% Success Rate
  Against the Strong Black-box Models of GPT-4.5/4o/o1
  Zhaoyi Li (MBZUAI)
  Xiaohan Zhao (MBZUAI)
  Dong-Dong Wu (The University of Tokyo/RIKEN AIP)
  Jiacheng Cui (MBZUAI)
  Zhiqiang Shen (MBZUAI)
• Approximation theory for 1-Lipschitz ResNets
  Davide Murari (University of Cambridge)
  Takashi Furuya (Doshisha University/RIKEN AIP)
  Carola-Bibiane Schönlieb (University of Cambridge)
• Bandit and Delayed Feedback in Online Structured Prediction
  Yuki Shibukawa (The University of Tokyo)
  Taira Tsuchiya (The University of Tokyo / RIKEN AIP)
  Shinsaku Sakaue (CyberAgent / The University of Tokyo / RIKEN AIP)
  Kenji Yamanishi (The University of Tokyo)
•  Balancing Positive and Negative Classification Error Rates in Positive-Unlabeled Learning
  Ximing Li (Jilin University / RIKEN AIP)
  Yuanchao Dai (Jilin University)
  Bing Wang (Jilin University)
  Changchun Li (Jilin University)
  Jianfeng Qu (Soochow University)
  Renchu Guan (Jilin University);
• Compact Memory for Continual Logistic Regression
  Yohan Jung (RIKEN AIP)
  Hyungi Lee* (KAIST)++
  Wenlong Chen* (Imperial College London)++
  Thomas Möllenhoff (RIKEN AIP)
  Yingzhen Li (Imperial College London)
  Juho Lee (KAIST)
  Mohammad Emtiyaz Khan (RIKEN AIP)
  *work done during internship at RIKEN AIP
• Degrees of Freedom for Linear Attention: Distilling Softmax Attention with Optimal Feature Efficiency
  Naoki Nishikawa (The University of Tokyo / RIKEN AIP)
  Rei Higuchi (The University of Tokyo / RIKEN AIP)
  Taiji Suzuki (The University of Tokyo / RIKEN AIP)
• DisasterM3: A Remote Sensing Vision-Language Dataset for Disaster Damage Assessment and Response
  Junjue Wang (The University of Tokyo)
  Weihao Xuan (The University of Tokyo/RIKEN AIP)
  Heli Qi (RIKEN AIP / Waseda University)
  Zhihao Liu (The University of Tokyo / RIKEN AIP)
  Kunyi Liu (Waseda University), Yuhan Wu (Stony Brook University)
  Hongruixuan Chen (The University of Tokyo / RIKEN AIP)
  Jian Song (RIKEN AIP)
  Junshi Xia (RIKEN AIP)
  Zhuo Zheng (Stanford University)
  Naoto Yokoya (The University of Tokyo/RIKEN AIP)
• Dr. RAW: Towards General High-Level Vision from RAW with Efficient Task Conditioning
  Wenjun Huang (University of California, Irvine)
  Ziteng Cui (The University of Tokyo)
  Yinqiang Zheng (The University of Tokyo)
  Yirui He (University of California, Irvine)
  Tatsuya Harada (The University of Tokyo / RIKEN AIP)
  Mohsen Imani (University of California, Irvine)
• DyG-Mamba: Continuous State Space Modeling on Dynamic Graphs
  Dongyuan Li (Tokyo Institute of Technology)
  Shiyin Tan (Tokyo Institute of Technology)
  Ying Zhang (RIKEN AIP)
  Ming Jin (Griffith University)
  Shirui Pan (Griffith University)
  Manabu Okumura (Tokyo Institute of Technology / RIKEN AIP)
  Renhe Jiang (The University of Tokyo)
• DynamicVL: Benchmarking Multimodal Large Language Models for Dynamic City Understanding
  Weihao Xuan (The University of Tokyo/RIKEN AIP)
  Junjue Wang (The University of Tokyo)
  Heli Qi (RIKEN AIP / Waseda University)
  Zihang Chen (Wuhan University)
  Zhuo Zheng (Stanford University)
  Yanfei Zhong (Wuhan University)
  Junshi Xia (RIKEN AIP)
  Naoto Yokoya (The University of Tokyo / RIKEN AIP)
• Efficient Low Rank Attention for Long-Context Inference in Large Language Models
  Tenghui Li+(Guangdong University of Technology / RIKEN AIP)
  Guoxu Zhou (Guangdong University of Technology)
  Xuyang Zhao (RIKEN iTHEMS / RIKEN IMS / Chiba University)
  Yuning Qiu (RIKEN AIP)
  Qibin Zhao (RIKEN AIP)
• Generalization Bound of Gradient Flow through Training Trajectory and Data-dependent Kernel
  Yilan Chen (University of California, San Diego)
  Zhichao Wang (University of California, Berkeley)
  Wei Huang (RIKEN AIP)
  Andi Han (University of Sydney)
  Taiji Suzuki (The University of Tokyo / RIKEN AIP)
  Arya Mazumdar (University of California / San Diego)
• Generalized Linear Bandits: Almost Optimal Regret with One-Pass Update
  Yu-Jie Zhang (RIKEN AIP)
  Sheng-An Xu (University of California, Berkeley)
  Peng Zhao (Nanjing University)
  Masashi Sugiyama (RIKEN AIP / The University of Tokyo)
• Graph Few-Shot Learning via Adaptive Spectrum Experts and Cross-Set Distribution Calibration
  Yonghao Liu (Jilin University)
  Yajun Wang (Jilin University)
  Chunli Guo (Jilin University)
  Wei Pang (Heriot-Watt University)
  Ximing Li (Jilin University / RIKEN AIP)
  Fausto Giunchiglia (Università di Trento)
  Xiaoyue Feng (Jilin University)
  Renchu Guan (Jilin University)
• Hessian-guided Perturbed Wasserstein Gradient Flows for Escaping Saddle
  Naoya Yamamoto (The University of Tokyo),
  Juno Kim (University of California, Berkeley / The University of Tokyo / RIKEN AIP),
  Taiji Suzuki (The University of Tokyo / RIKEN AIP)
• How Does Label Noise Gradient Descent Improve Generalization in the Low SNR Regime?
  Wei Huang (RIKEN AIP)
  Andi Han (University of Sydney)
  Yujin Song (Exawizards Inc.)
  Yilan Chen (University of California, San Diego)
  Denny Wu (New York University / Flatiron Institute)
  Difan Zou (University of Hong Kong)
  Taiji Suzuki (The University of Tokyo / RIKEN AIP)
• Infinite-Width Limit of a Single Attention Layer: Analysis via Tensor Programs
  Mana Sakai (The University of Tokyo / RIKEN AIP)
  Ryo Karakida (The National Institute of Advanced Industrial Science and Technology)
  Masaaki Imaizumi (The University of Tokyo / RIKEN AIP)
• Information-theoretic Generalization Analysis for VQ-VAEs: A Role of Latent Variables
  Futoshi Futami* (The University of Osaka/ RIKEN AIP)
  Masahiro Fujisawa* (The University of Osaka/ RIKEN AIP)
  * Equal contribution
• Intend to Move: A Dataset for Intention and Scene Aware Human Motion Prediction
  Ryo Umagami (The University of Tokyo)
  Liu Yue (The University of Tokyo)
  Xuangeng Chu (The University of Tokyo)
  Ryuto Fukushima (The University of Tokyo)
  Tetsuya Narita (The University of Tokyo)
  Yusuke Mukuta (The University of Tokyo)
  Tomoyuki Takahata (Tokyo Denki University)
  Jianfei Yang (Nanyang Technological University)
  Tatsuya Harada (The University of Tokyo / RIKEN AIP)
• I2-NeRF: Learning Neural Radiance Fields Under Physically-Grounded Media Interactions
  Shuhong Liu (The University of Tokyo)
  Lin Gu (RIKEN AIP)
  Ziteng Cui (The University of Tokyo)
  Xuangeng Chu (The University of Tokyo)
  Tatsuya Harada (The University of Tokyo / RIKEN AIP)
• Last Iterate Convergence in Monotone Mean Field Games
  Noboru Isobe (RIKEN AIP)
  Kenshi Abe (Cyber Agent)
  Kaito Ariu (Cyber Agent)
• Neural Collapse in Cumulative Link Models for Ordinal Regression: An Analysis with Unconstrained Feature Model
  Chuang Ma (Kyoto University)
  Tomoyuki Obuchi (Kyoto University / RIKEN AIP)
  Toshiyuki Tanaka (Kyoto University)
• Online Inverse Linear Optimization: Efficient Logarithmic-Regret Algorithm, Robustness to Suboptimality, and Lower Bound
  Shinsaku Sakaue (CyberAgent / The University of Tokyo / RIKEN AIP)
  Taira Tsuchiya (The University of Tokyo / RIKEN AIP)
  Han Bao (Institute of Statistical Mathematics)
  Taihei Oki (Hokkaido University / RIKEN AIP)
• Optimal Dynamic Regret by Transformers for Non-Stationary Reinforcement Learning
  Baiyuan Chen (The University of Tokyo)
  Shinji Ito (The University of Tokyo / RIKEN AIP)
  Masaaki Imaizumi (The University of Tokyo / RIKEN AIP)
• Optimal Estimation of the Best Mean in Multi-Armed Bandits
  Takayuki Osogami (IBM)
  Junya Honda (Kyoto University / RIKEN AIP)
  Junpei Komiyama (New York University / RIKEN AIP)
• Optimal Regret of Bandits under Differential Privacy
  Achraf Azize (FairPlay Joint Team, CREST, ENSAE Paris)
  Yulian Wu* (King Abdullah University of Science and Technology)++
  Junya Honda (Kyoto University / RIKEN AIP)
  Francesco Orabona (King Abdullah University of Science and Technology)
  Shinji Ito (The University of Tokyo / RIKEN AIP)
  Debabrota Basu (Univ. Lille, Inria, CNRS, Centrale Lille)
  *work done during internship at RIKEN AIP
• Quantifying Statistical Significance of Deep Nearest Neighbor Anomaly Detection via Selective Inference.
  Mizuki Niihori (Nagoya University)
  Shuichi Nishino (Nagoya University / RIKEN AIP)
  Teruyuki Katsuoka (Nagoya University)
  Tomohiro Shiraishi (Nagoya University / RIKEN AIP)
  Ichiro Takeuchi (Nagoya University / RIKEN AIP)
• Quantization Error Propagation: Revisiting Layer-Wise Post-Training Quantization
  Yamato Arai (Fujitsu Limited / The University of Tokyo)
  Yuma Ichikawa (Fujitsu Limited / RIKEN AIP)
• Revisiting Follow-the-Perturbed-Leader with Unbounded Perturbations in Bandit Problems
  Jongyeong Lee (Seoul National University)
  Junya Honda (Kyoto University / RIKEN AIP)
  Shinji Ito (The University of Tokyo / RIKEN AIP)
  Min-hwan Oh (Seoul National University)
• Scalable Valuation of Human Feedback through Provably Robust Model Alignment
  Masahiro Fujisawa* (The University of Osaka / RIKEN AIP)
  Masaki Adachi* (Toyota Motor Corporation / University of Oxford)
  Michael A. Osborne (University of Oxford)
  *Equal contribution
• State Size Independent Statistical Error Bound for Discrete Diffusion Models
  Shintaro Wakasugi (The University of Tokyo),
  Taiji Suzuki (The University of Tokyo / RIKEN AIP).
• The adaptive complexity of minimizing relative fisher information
  Huanjian Zhou (The University of Tokyo / RIKEN AIP)
  Masashi Sugiyama (RIKEN/The University of Tokyo)
• The Effect of Optimal Self-Distillation in Noisy Gaussian Mixture Model
  Kaito Takanami (The University of Tokyo / Ochanomizu University)
  Takashi Takahashi (The University of Tokyo / RIKEN AIP)
  Ayaka Sakata (Ochanomizu University / RIKEN AIP)
• Towards a Geometric Understanding of Tensor Learning via the t-Product
  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)
• Trained Mamba Emulates Online Gradient Descent in In-Context Linear Regression
  Jiarui Jiang (Harbin Institute of Technology, Shenzhen)
  Wei Huang (RIKEN AIP)
  Miao Zhang (Harbin Institute of Technology, Shenzhen)
  Taiji Suzuki (The University of Tokyo / RIKEN AIP)
  Liqiang Nie (Harbin Institute of Technology, Shenzhen)
• UEPI: Universal Energy-Behavior-Preserving Integrators for
  Energy Conservative/Dissipative Differential Equations
  Elena Celledoni(NTNU)
  Brynjulf Owren(NTNU)
  Chong Shen(Kobe University)
  Baige Xu(Kobe University)
  Takaharu Yaguchi(Kobe University / RIKEN AIP)
• Wavy Transformer
  Satoshi Noguchi (JAMSTEC / RIKEN AIP)
  Yoshinobu Kawahara (The University of Osaka / RIKEN AIP)

Creative AI Track, Artworks

• Melody Slot Machine III for Automatic Fingering Saxophone using Servomotors
  Masatoshi Hamanaka (RIKEN AIP)
  Gou Koutaki（Kumamoto University）

+Past Student Trainee
++Past Intern

Related Laboratories