NeurIPS

NaViL: Rethinking Scaling Properties of Native Multimodal Large Language Models under Data Constraints

Changyao Tian, Hao Li, Gen Luo, Xizhou Zhu, Weijie Su, Hanming Deng, Jinguo Zhu, Jie Shao, Ziran Zhu, Yunpeng Liu, Lewei Lu, Wenhai Wang, Hongsheng Li, Jifeng Dai

机构 * Shanghai AI Laboratory（上海人工智能实验室） ； The Chinese University of Hong Kong（香港中文大学） ； Tsinghua University（清华大学） ； Sensetime Research（商汤科技研究院） ； Nanjing University（南京大学） 纠错

Comments Accepted by NeurIPS 2025. 22 pages, link: https://github.com/OpenGVLab/NaViL

Convergence Theorems for Entropy-Regularized and Distributional Reinforcement Learning

Yash Jhaveri, Harley Wiltzer, Patrick Shafto, Marc G. Bellemare, David Meger

机构 * Rutgers University–Newark（罗格斯大学-新布朗斯维尔分校） ； Mila–Québec AI Institute（魁北克人工智能研究所） ； McGill University（麦吉尔大学） 纠错

Comments Accepted to NeurIPS 2025. First two authors contributed equally

Dynamic Features Adaptation in Networking: Toward Flexible training and Explainable inference

Yannis Belkhiter, Seshu Tirupathi, Giulio Zizzo, Merim Dzaferagic, John D. Kelleher

机构 * IBM Research Europe（IBM欧洲研究院） ； Trinity College Dublin（三一学院） ； ADAPT Research Centre（ADAPT研究中心） 纠错

Comments Accepted at AI4NextG Workshop, NeurIPS 2025

Approximate Domain Unlearning for Vision-Language Models

Kodai Kawamura, Yuta Goto, Rintaro Yanagi, Hirokatsu Kataoka, Go Irie

机构 * Tokyo University of Science（东京科学大学） ； National University of Singapore（新加坡国立大学） ； National Institute of Advanced Industrial Science and Technology (AIST)（国家先进工业科学与技术研究院） ； University of Oxford（牛津大学） 纠错

Comments NeurIPS 2025 (Spotlight)

Efficient Kernelized Learning in Polyhedral Games Beyond Full-Information: From Colonel Blotto to Congestion Games

Andreas Kontogiannis, Vasilis Pollatos, Gabriele Farina, Panayotis Mertikopoulos, Ioannis Panageas

Comments Accepted at NeurIPS 2025

DiffEye: Diffusion-Based Continuous Eye-Tracking Data Generation Conditioned on Natural Images

Ozgur Kara, Harris Nisar, James M. Rehg

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

Comments Accepted to NeurIPS 2025

Training a Foundation Model for Materials on a Budget

Teddy Koker, Mit Kotak, Tess Smidt

机构 * Department of Electrical Engineering and Computer Science（电气工程与计算机科学系） 纠错

Comments To appear at the NeurIPS 2025 Workshop on AI for Accelerated Materials Discovery

Feedback Guidance of Diffusion Models

Felix Koulischer, Florian Handke, Johannes Deleu, Thomas Demeester, Luca Ambrogioni

机构 * Ghent University - imec（根特大学 - imec） ； Donders Institute for Brain Cognition and Behaviour（大脑认知与行为研究所） 纠错

Comments Article accepeted as poster at the 39th Annual Conference on Neural Information Processing Systems (NeurIPS25). Code is available at: https://github.com/FelixKoulischer/FBG_using_edm2

Learning When to Think: Shaping Adaptive Reasoning in R1-Style Models via Multi-Stage RL

Songjun Tu, Jiahao Lin, Qichao Zhang, Xiangyu Tian, Linjing Li, Xiangyuan Lan, Dongbin Zhao

机构 * Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所） ； Pengcheng Laboratory（鹏城实验室） ； School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院） 纠错

Journal ref NeurIPS 2025

Multi-Turn Human-LLM Interaction Through the Lens of a Two-Way Intelligibility Protocol

Harshvardhan Mestha, Karan Bania, Shreyas V Sathyanarayana, Sidong Liu, Ashwin Srinivasan

机构 * Dept. of Electrical and Electronics Engineering（电子工程系） ； BITS Pilani, K.K. Birla Goa Campus（比斯汉学院，戈阿校园） ； Machine Learning Department（机器学习系） ； Carnegie Mellon University（卡内基梅隆大学） ； Centre for Health Informatics（健康信息学中心） ； Macquarie University, Sydney（麦考瑞大学，悉尼） ； Dept. of Computer Science & Information Systems（计算机科学与信息系统系） 纠错

Comments Multi-Turn Interactions in Large Language Models (MTI-LLM) Workshop at NeurIPS 2025

DemandCast: Global hourly electricity demand forecasting

Kevin Steijn, Vamsi Priya Goli, Enrico Antonini

机构 * Open Energy Transition（开放能源转型） 纠错

Comments 7 pages, 4 figures, accepted at the NeurIPS 2025 Workshop: Tackling Climate Change with Machine Learning

Unveiling the Power of Multiple Gossip Steps: A Stability-Based Generalization Analysis in Decentralized Training

Qinglun Li, Yingqi Liu, Miao Zhang, Xiaochun Cao, Quanjun Yin, Li Shen

机构 * State Key Laboratory of Digital Intelligent Modeling and Simulation（数字智能建模与仿真国家重点实验室） ； National University of Defense Technology（国防科技大学） ； School of Cyber Science and Technology（网络安全科学与技术学院） ； Shenzhen Campus of Sun Yat-sen University（中山大学深圳校区） 纠错

Comments This paper has been accepted by NeurIPS 2025 (Spotlight)

PRESCRIBE: Predicting Single-Cell Responses with Bayesian Estimation

Jiabei Cheng, Changxi Chi, Jingbo Zhou, Hongyi Xin, Jun Xia

机构 * Shanghai Jiao Tong University（上海交通大学） ； Westlake University（西湖大学） ； AIMS Lab, The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州）AIMS实验室） ； The Hong Kong University of Science and Technology（香港科技大学） 纠错

Comments 39th Conference on Neural Information Processing Systems (NeurIPS 2025)

Synergy Between the Strong and the Weak: Spiking Neural Networks are Inherently Self-Distillers

Yongqi Ding, Lin Zuo, Mengmeng Jing, Kunshan Yang, Pei He, Tonglan Xie

机构 * School of Information and Software Engineering University of Electronic Science and Technology of China（信息与软件工程学院 电子科技大学） 纠错

Comments Accepted by NeurIPS 2025

MetaDefense: Defending Finetuning-based Jailbreak Attack Before and During Generation

Weisen Jiang, Sinno Jialin Pan

Comments Accepted By NeurIPS 2025

SIMU: Selective Influence Machine Unlearning

Anu Agarwal, Mihir Pamnani, Dilek Hakkani-Tur

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

Comments Accepted to NeurIPS 2025 Workshop: Constrained Optimization for Machine Learning (COML)

Rectified-CFG++ for Flow Based Models

Shreshth Saini, Shashank Gupta, Alan C. Bovik

机构 * The University of Texas at Austin（德克萨斯大学奥斯汀分校） 纠错

Comments Accepted at NeurIPS 2025

Evaluating and Learning Optimal Dynamic Treatment Regimes under Truncation by Death

Sihyung Park, Wenbin Lu, Shu Yang

Comments 30 pages, 5 figures, 6 tables, The Thirty-Ninth Annual Conference on Neural Information Processing Systems

ExpertAgent: Enhancing Personalized Education through Dynamic Planning and Retrieval-Augmented Long-Chain Reasoning

Binrong Zhu, Guiran Liu, Nina Jiang

机构 * Department of Computer Science（计算机科学系） ； San Francisco State University（旧金山州立大学） ； Department of Mechanical & Industrial Engineering（机械与工业工程系） ； Louisiana State University（路易斯安那州立大学） 纠错

Comments Manuscript previously submitted to the NeurIPS 2025 Workshop on Bridging Language, Agent, and World Models (LAW 2025)

Simulation-based inference for neutrino interaction model parameter tuning

Karla Tame-Narvaez, Aleksandra Ćiprijanović, Steven Gardiner, Giuseppe Cerati

Comments Accepted contribution to NeurIPS 2025 Workshop on Machine Learning and the Physical Sciences. 7 pages, 1 figure

MacroBench: A Novel Testbed for Web Automation Scripts via Large Language Models

Hyunjun Kim, Sejong Kim

机构 * KAIST（韩国科学技术院） 纠错

Comments NeurIPS 2025 Workshop on Lock-LLM

Speculate Deep and Accurate: Lossless and Training-Free Acceleration for Offloaded LLMs via Substitute Speculative Decoding

Pei-Shuo Wang, Jian-Jia Chen, Chun-Che Yang, Chi-Chih Chang, Ning-Chi Huang, Mohamed S. Abdelfattah, Kai-Chiang Wu

Comments Accepted by NeurIPS 2025

DNA-DetectLLM: Unveiling AI-Generated Text via a DNA-Inspired Mutation-Repair Paradigm

Xiaowei Zhu, Yubing Ren, Fang Fang, Qingfeng Tan, Shi Wang, Yanan Cao

机构 * Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所） ； School of Cyber Security, University of Chinese Academy of Sciences（中国科学院大学网络安全学院） ； University International College, Macau University of Science and Technology（澳门科技大学国际学院） ； Cyberspace Institute of Advanced Technology, Guangzhou University（广州大学网络安全技术研究所） ； Institute of Computing Science, Chinese Academy of Sciences（中国科学院计算科学研究所） 纠错

Comments NeurIPS 2025 Spotlight

FireGNN: Neuro-Symbolic Graph Neural Networks with Trainable Fuzzy Rules for Interpretable Medical Image Classification

Prajit Sengupta, Islem Rekik

机构 * Imperial College London（伦敦帝国学院） 纠错

Comments Accepted at NeurIPS 2025 Conference (Workshop Track), San Diego, USA

X-Teaming Evolutionary M2S: Automated Discovery of Multi-turn to Single-turn Jailbreak Templates

Hyunjun Kim, Junwoo Ha, Sangyoon Yu, Haon Park

机构 * AIM Intelligence（AIM智能） 纠错

Comments NeurIPS 2025 Workshop on Lock-LLM

Adaptive Originality Filtering: Rejection Based Prompting and RiddleScore for Culturally Grounded Multilingual Riddle Generation

Duy Le, Kent Ziti, Evan Girard-Sun, Bakr Bouhaya, Sean O'Brien, Vasu Sharma, Kevin Zhu

机构 * Algoverse AI Research（Algoverse AI研究院） 纠错

Comments Paper was accepted in to NeurIPS 2025 Workshop GenProCC

ObjexMT: Objective Extraction and Metacognitive Calibration for LLM-as-a-Judge under Multi-Turn Jailbreaks

Hyunjun Kim, Junwoo Ha, Sangyoon Yu, Haon Park

Comments NeurIPS 2025 Workshop on MTI-LLM

Martingale Posterior Neural Networks for Fast Sequential Decision Making

Gerardo Duran-Martin, Leandro Sánchez-Betancourt, Álvaro Cartea, Kevin Murphy

机构 * Oxford-Man Institute of Quantitative Finance（牛津-曼定量金融研究所） ； Mathematical Institute, University of Oxford（牛津大学数学研究所） ； Google Deepmind（谷歌DeepMind） 纠错

Journal ref Advances in Neural Information Processing Systems 39 (NeurIPS 2025)

Trans-EnV: A Framework for Evaluating the Linguistic Robustness of LLMs Against English Varieties

Jiyoung Lee, Seungho Kim, Jieun Han, Jun-Min Lee, Kitaek Kim, Alice Oh, Edward Choi

机构 * KAIST（韩国科学技术院） ； Suresoft Technologies（Suresoft技术公司） ； Seoul National University（首尔国立大学） 纠错

Comments NeurIPS 2025 Track on Datasets and Benchmarks (27 pages, 6 figures, 16 tables)

SQL-R1: Training Natural Language to SQL Reasoning Model By Reinforcement Learning

Peixian Ma, Xialie Zhuang, Chengjin Xu, Xuhui Jiang, Ran Chen, Jian Guo

Comments Accepd by NeurIPS 2025