NeurIPS

Mesh-RFT: Enhancing Mesh Generation via Fine-grained Reinforcement Fine-Tuning

Jian Liu, Jing Xu, Song Guo, Jing Li, Jingfeng Guo, Jiaao Yu, Haohan Weng, Biwen Lei, Xianghui Yang, Zhuo Chen, Fangqi Zhu, Tao Han, Chunchao Guo

机构 * Hong Kong University of Science and Technology（香港科技大学） ； Tencent Hunyuan（腾讯文元） ； University of Science and Technology of China（中国科学技术大学） ； South China University of Technology（华南理工大学） 纠错

Comments NeurIPS 2025, Spotlight

RL Tango: Reinforcing Generator and Verifier Together for Language Reasoning

Kaiwen Zha, Zhengqi Gao, Maohao Shen, Zhang-Wei Hong, Duane S. Boning, Dina Katabi

机构 * MIT（麻省理工学院） ； MIT-IBM Watson AI Lab（麻省理工-IBM沃森人工智能实验室） 纠错

Comments NeurIPS 2025. The first two authors contributed equally

SAFEPATH: Preventing Harmful Reasoning in Chain-of-Thought via Early Alignment

Wonje Jeung, Sangyeon Yoon, Minsuk Kahng, Albert No

机构 * Department of Artificial Intelligence, Yonsei University（人工智能系，延世大学） ； Department of Computer Science and Engineering, Yonsei University（计算机科学与工程系，延世大学） 纠错

Comments Accepted at NeurIPS 2025. Code and models are available at https://ai-isl.github.io/safepath

CALM-PDE: Continuous and Adaptive Convolutions for Latent Space Modeling of Time-dependent PDEs

Jan Hagnberger, Daniel Musekamp, Mathias Niepert

机构 * Machine Learning and Simulation Lab（机器学习与仿真实验室） ； Institute for Artificial Intelligence（人工智能研究所） ； University of Stuttgart（斯图加特大学） ； International Max Planck Research School for Intelligent Systems (IMPRS-IS)（智能系统国际马克斯·普朗克研究学校） ； Stuttgart Center for Simulation Science (SimTech)（斯图加特模拟科学中心） 纠错

Comments Accepted for publication at the 39th Conference on Neural Information Processing Systems (NeurIPS) 2025, San Diego, California, USA

Shallow Flow Matching for Coarse-to-Fine Text-to-Speech Synthesis

Dong Yang, Yiyi Cai, Yuki Saito, Lixu Wang, Hiroshi Saruwatari

机构 * The University of Tokyo（东京大学） ； Nanyang Technological University（南洋理工大学） 纠错

Comments Accepted by NeurIPS 2025

Learning Dense Hand Contact Estimation from Imbalanced Data

Daniel Sungho Jung, Kyoung Mu Lee

机构 * IPAI ； Dept. of ECE & ASRI（电子工程与先进机器人研究所） 纠错

Comments Accepted at NeurIPS 2025. Project page: http://haco-release.github.io

UMoE: Unifying Attention and FFN with Shared Experts

Yuanhang Yang, Chaozheng Wang, Jing Li

机构 * Institute of Science Tokyo（东京科学研究所） ； The Chinese University of Hong Kong（香港中文大学） ； Hong Kong Polytechnic University（香港理工大学） 纠错

Comments NeurIPS 2025 Spotlight

Stop Summation: Min-Form Credit Assignment Is All Process Reward Model Needs for Reasoning

Jie Cheng, Gang Xiong, Ruixi Qiao, Lijun Li, Chao Guo, Junle Wang, Yisheng Lv, Fei-Yue Wang

机构 * State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences（多模态人工智能系统国家重点实验室，自动化研究所，中国科学院） ； School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院） ； Shanghai Artificial Intelligence Laboratory（上海人工智能实验室） ； Tencent（腾讯） ； Faculty of Innovation Engineering, Macau University of Science and Technology（澳门科技大学创新工程学院） ； DeSci Center of Parallel Intelligence, Obuda University（并行智能德Sci中心，奥布达大学） ； Research Center for Chinese Economics and Social Security, University of Chinese Academy of Sciences（中国经济社会安全研究中心，中国科学院大学） ； Institute of Automation, Chinese Academy of Sciences（自动化研究所，中国科学院） 纠错

Comments Accepted by NeurIPS 2025

Sign-In to the Lottery: Reparameterizing Sparse Training From Scratch

Advait Gadhikar, Tom Jacobs, Chao Zhou, Rebekka Burkholz

机构 * CISPA Helmholtz Center for Information Security（CISPA赫尔姆霍茨信息 security 研究所） 纠错

Comments Accepted at NeurIPS 2025

Streaming Federated Learning with Markovian Data

Tan-Khiem Huynh, Malcolm Egan, Giovanni Neglia, Jean-Marie Gorce

机构 * Inria（法国国家信息与自动化技术研究所） ； INSA Lyon（里昂国立应用科学学院） ； Université Côte d’Azur（蔚蓝海岸大学） 纠错

Comments Neurips 2025 camera-ready version

Training Robust Graph Neural Networks by Modeling Noise Dependencies

Yeonjun In, Kanghoon Yoon, Sukwon Yun, Kibum Kim, Sungchul Kim, Chanyoung Park

机构 * KAIST（韩国科学技术院） ； UNC Chapel Hill（北卡罗来纳大学教堂山分校） ； Adobe Research（Adobe研究） 纠错

Comments NeurIPS 2025

Harnessing Feature Resonance under Arbitrary Target Alignment for Out-of-Distribution Node Detection

Shenzhi Yang, Junbo Zhao, Sharon Li, Shouqing Yang, Dingyu Yang, Xiaofang Zhang, Haobo Wang

机构 * Zhejiang University（浙江大学） ； Hangzhou High-Tech Zone (Binjiang) Institute of Blockchain and Data Security（杭州高新技术区（滨江）区块链与数据安全研究院） ； Department of Computer Sciences, University of Wisconsin-Madison（威斯康星大学麦迪逊分校计算机科学系） ； School of Computer Science and Technology, Soochow University（苏州大学计算机科学与技术学院） 纠错

Comments NeurIPS 2025 Accepted

Continuous Diffusion Model for Language Modeling

Jaehyeong Jo, Sung Ju Hwang

机构 * KAIST（韩国科学技术院） 纠错

Comments NeurIPS 2025

MIR-Bench: Can Your LLM Recognize Complicated Patterns via Many-Shot In-Context Reasoning?

Kai Yan, Zhan Ling, Kang Liu, Yifan Yang, Ting-Han Fan, Lingfeng Shen, Zhengyin Du, Jiecao Chen

机构 * University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校） 纠错

Comments 39 pages, 11 figures. The paper is accepted at NeurIPS 2025 Datasets & Benchmarks Track, and the latest version adds modifications in camera-ready

Depth-Bounds for Neural Networks via the Braid Arrangement

Moritz Grillo, Christoph Hertrich, Georg Loho

机构 * Max Planck Institute for Mathematics in the Sciences（马克斯·普朗克数学研究所） ； University of Technology Nuremberg（纽伦堡技术大学） ； Freie Universität Berlin（柏林自由大学） ； University of Twente（埃因霍温理工大学） 纠错

Comments Accepted at NeurIPS 2025

SMRS: advocating a unified reporting standard for surrogate models in the artificial intelligence era

Elizaveta Semenova, Alisa Sheinkman, Timothy James Hitge, Siobhan Mackenzie Hall, Jon Cockayne

机构 * Imperial College London（伦敦帝国学院） ； University of Edinburgh（爱丁堡大学） ； AIMS South Africa（南非AIMS研究所） ； University of Oxford（牛津大学） ； University of Southampton（南安普顿大学） 纠错

Comments Accepted at the 39th Conference on Neural Information Processing Systems (NeurIPS 2025), Position Track

Heterogeneous Swarms: Jointly Optimizing Model Roles and Weights for Multi-LLM Systems

Shangbin Feng, Zifeng Wang, Palash Goyal, Yike Wang, Weijia Shi, Huang Xia, Hamid Palangi, Luke Zettlemoyer, Yulia Tsvetkov, Chen-Yu Lee, Tomas Pfister

机构 * University of Washington（华盛顿大学） ； Google Cloud AI Research（谷歌云人工智能研究） 纠错

Comments NeurIPS 2025

Face-Human-Bench: A Comprehensive Benchmark of Face and Human Understanding for Multi-modal Assistants

Lixiong Qin, Shilong Ou, Miaoxuan Zhang, Jiangning Wei, Yuhang Zhang, Xiaoshuai Song, Yuchen Liu, Mei Wang, Weiran Xu

机构 * Beijing University of Posts and Telecommunications（北京邮电大学） ； Beijing Normal University（北京师范大学） 纠错

Comments 50 pages, 14 figures, 42 tables. NeurIPS 2025 Datasets and Benchmarks Track

Deep Continuous-Time State-Space Models for Marked Event Sequences

Yuxin Chang, Alex Boyd, Cao Xiao, Taha Kass-Hout, Parminder Bhatia, Padhraic Smyth, Andrew Warrington

机构 * University of California, Irvine（加州大学尔湾分校） ； GE HealthCare（通用电气医疗） 纠错

Comments NeurIPS 2025 Spotlight

JAMUN: Bridging Smoothed Molecular Dynamics and Score-Based Learning for Conformational Ensembles

Ameya Daigavane, Bodhi P. Vani, Darcy Davidson, Saeed Saremi, Joshua Rackers, Joseph Kleinhenz

Comments 37 pages, accepted at NeurIPS 2025

Temporal-Difference Variational Continual Learning

Luckeciano C. Melo, Alessandro Abate, Yarin Gal

机构 * OATML, University of Oxford（OATML，牛津大学） ； OXCAV, University of Oxford（OXCAV，牛津大学） 纠错

Comments Published at NeurIPS 2025

Towards Understanding Safety Alignment: A Mechanistic Perspective from Safety Neurons

Jianhui Chen, Xiaozhi Wang, Zijun Yao, Yushi Bai, Lei Hou, Juanzi Li

机构 * Department of Computer Science and Technology, BNRist（计算机科学与技术系，BNRist） ； Shenzhen International Graduate School（深圳国际研究生院） ； KIRC, Institute for Artificial Intelligence, Tsinghua University（人工智能研究院，清华大学） 纠错

Comments NeurIPS 2025

Abstain Mask Retain Core: Time Series Prediction by Adaptive Masking Loss with Representation Consistency

Renzhao Liang, Sizhe Xu, Chenggang Xie, Jingru Chen, Feiyang Ren, Shu Yang, Takahiro Yabe

机构 * Beihang University（北航） ； New York University（纽约大学） ； Peking University（北京大学） ； University of Leeds（利兹大学） 纠错

Comments 20 pages, 4 figures. Accepted as Spotlight poster in NeurIPS 2025

Are Greedy Task Orderings Better Than Random in Continual Linear Regression?

Matan Tsipory, Ran Levinstein, Itay Evron, Mark Kong, Deanna Needell, Daniel Soudry

Comments Accepted to NeurIPS 2025

Mitigating Privacy-Utility Trade-off in Decentralized Federated Learning via $f$-Differential Privacy

Xiang Li, Buxin Su, Chendi Wang, Qi Long, Weijie J. Su

机构 * University of Pennsylvania（宾夕法尼亚大学） ； Xiamen University（厦门大学） 纠错

Comments NeurIPS 2025 (Spotlight)

Compressing Biology: Evaluating the Stable Diffusion VAE for Phenotypic Drug Discovery

Télio Cropsal, Rocío Mercado

机构 * Chalmers University of Technology and University of Gothenburg（查尔姆斯理工大学和哥德堡大学） 纠错

Comments Accepted to the 3rd Workshop on Imageomics: Discovering Biological Knowledge from Images Using AI at the 39th Conference on Neural Information Processing Systems (NeurIPS 2025)

Multi-Resolution Analysis of the Convective Structure of Tropical Cyclones for Short-Term Intensity Guidance

Elizabeth Cucuzzella, Tria McNeely, Kimberly Wood, Ann B. Lee

机构 * Department of Statistics and Data Science（统计与数据科学系） ； Carnegie Mellon University（卡内基梅隆大学） ； Department of Atmospheric Sciences and Hydrology（大气科学与水文学系） ； University of Arizona（亚利桑那大学） 纠错

Comments For Tackling Climate Change with Machine Learning workshop at NeurIPS 2025

Deep Research Brings Deeper Harm

Shuo Chen, Zonggen Li, Zhen Han, Bailan He, Tong Liu, Haokun Chen, Georg Groh, Philip Torr, Volker Tresp, Jindong Gu

机构 * LMU Munich（慕尼黑大学） ； Siemens（西门子） ； Munich Center for Machine Learning (MCML)（慕尼黑机器学习中心） ； Technical University of Munich (TUM)（慕尼黑技术大学） ； AWS AI（亚马逊AI） ； Konrad Zuse School of Excellence in Reliable AI (relAI)（Konrad Zuse卓越可靠AI学校） ； University of Hong Kong (HKU)（香港大学） ； University of Oxford（牛津大学） 纠错

Comments Accepted to Reliable ML from Unreliable Data Workshop @ NeurIPS 2025

What Happens During the Loss Plateau? Understanding Abrupt Learning in Transformers

Pulkit Gopalani, Wei Hu

机构 * University of Michigan（密歇根大学） 纠错

Comments NeurIPS 2025

Roboflow100-VL: A Multi-Domain Object Detection Benchmark for Vision-Language Models

Peter Robicheaux, Matvei Popov, Anish Madan, Isaac Robinson, Joseph Nelson, Deva Ramanan, Neehar Peri

机构 * Roboflow ； Carnegie Mellon University（卡内基梅隆大学） 纠错

Comments The first two authors contributed equally. This work has been accepted to the Neural Information Processing Systems (NeurIPS) 2025 Datasets & Benchmark Track. Project Page: https://rf100-vl.org/