NeurIPS

GASP: Efficient Black-Box Generation of Adversarial Suffixes for Jailbreaking LLMs

Advik Raj Basani, Xiao Zhang

机构 * Birla Institute of Technology and Science（比拉理工学院） ； CISPA Helmholtz Center for Information Security（信息安全赫尔姆霍茨中心） 纠错

Comments Accepted to NeurIPS 2025. Project page and demos: https://air-ml.org/project/gasp/

Understanding Adam Requires Better Rotation Dependent Assumptions

Tianyue H. Zhang, Lucas Maes, Alan Milligan, Alexia Jolicoeur-Martineau, Ioannis Mitliagkas, Damien Scieur, Simon Lacoste-Julien, Charles Guille-Escuret

机构 * Mila, Quebec AI Institute（魁北克人工智能研究所） ； Université de Montréal（蒙特利尔大学） ； University of British Columbia（不列颠哥伦比亚大学） ； Samsung SAIL Montreal（三星SAIL蒙特利尔分校） ； Canada CIFAR AI Chair（加拿大CIFAR人工智能主席） ； Archimedes Unit, Athena Research Center（阿提卡研究中心阿基米德单位） 纠错

Comments Published at NeurIPS 2025

Text to Sketch Generation with Multi-Styles

Tengjie Li, Shikui Tu, Lei Xu

机构 * School of Computer Science, Shanghai Jiao Tong University（上海交通大学计算机科学学院） ； Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ)（广东省人工智能与数字经济实验室（深圳）） 纠错

Comments Accepted by NeurIPS 2025

Tortoise and Hare Guidance: Accelerating Diffusion Model Inference with Multirate Integration

Yunghee Lee, Byeonghyun Pak, Junwha Hong, Hoseong Kim

Comments 21 pages, 8 figures. NeurIPS 2025. Project page: https://yhlee-add.github.io/THG

DartQuant: Efficient Rotational Distribution Calibration for LLM Quantization

Yuantian Shao, Yuanteng Chen, Peisong Wang, Jianlin Yu, Jing Lin, Yiwu Yao, Zhihui Wei, Jian Cheng

机构 * Nanjing University of Science and Technology（南京理工大学） ； School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院） ； Huawei Technologies Co., Ltd.（华为技术有限公司） 纠错

Comments NeurIPS 2025, 10 pages, 12 figures

Towards Scalable Meta-Learning of near-optimal Interpretable Models via Synthetic Model Generations

Kyaw Hpone Myint, Zhe Wu, Alexandre G. R. Day, Giri Iyengar

机构 * Capital One 纠错

Comments 9 pages, 3 figures, Neurips 2025 GenAI in Finance Workshop

Context informs pragmatic interpretation in vision-language models

Alvin Wei Ming Tan, Ben Prystawski, Veronica Boyce, Michael C. Frank

机构 * Department of Psychology Stanford University（心理学系 斯坦福大学） 纠错

Comments Accepted at CogInterp Workshop, NeurIPS 2025

To See or To Read: User Behavior Reasoning in Multimodal LLMs

Tianning Dong, Luyi Ma, Varun Vasudevan, Jason Cho, Sushant Kumar, Kannan Achan

机构 * Personalization Team, Walmart Global Tech（Walmart全球科技个性化团队） 纠错

Comments Accepted by the 39th Conference on Neural Information Processing Systems (NeurIPS 2025) Workshop: Efficient Reasoning

ASAP: an Agentic Solution to Auto-optimize Performance of Large-Scale LLM Training

Yuran Ding, Xinwei Chen, Xiaofan Zhang, Zongwei Zhou

Comments This work has been accepted to Workshop on ML for Systems at NeurIPS 2025

Sketch-Augmented Features Improve Learning Long-Range Dependencies in Graph Neural Networks

Ryien Hosseini, Filippo Simini, Venkatram Vishwanath, Rebecca Willett, Henry Hoffmann

机构 * Department of Computer Science, University of Chicago（芝加哥大学计算机科学系） ； Leadership Computing Facility, Argonne National Laboratory（阿贡国家实验室领导计算设施） ； Department of Statistics, University of Chicago（芝加哥大学统计系） ； NSF-Simons National Institute for Theory and Mathematics in Biology（国家理论与生物数学研究所） 纠错

Comments To appear at NeurIPS 2025

Optimizing Reasoning Efficiency through Prompt Difficulty Prediction

Bo Zhao, Berkcan Kapusuzoglu, Kartik Balasubramaniam, Sambit Sahu, Supriyo Chakraborty, Genta Indra Winata

机构 * UC San Diego（加州大学圣地亚哥分校） ； Capital One（Capital One公司） 纠错

Comments NeurIPS 2025 Workshop on Efficient Reasoning

Learning Paths for Dynamic Measure Transport: A Control Perspective

Aimee Maurais, Bamdad Hosseini, Youssef Marzouk

机构 * Massachusetts Institute of Technology（麻省理工学院） ； University of Washington（华盛顿大学） 纠错

Comments To appear at NeurIPS 2025 Workshop on Frontiers of Probabilistic Inference: Sampling Meets Learning

What's in Common? Multimodal Models Hallucinate When Reasoning Across Scenes

Candace Ross, Florian Bordes, Adina Williams, Polina Kirichenko, Mark Ibrahim

机构 * FAIR at Meta（Meta 的 FAIR） 纠错

Comments 10 pages, 6 figures. Accepted to NeurIPS Datasets & Benchmarks 2025

Node-Based Editing for Multimodal Generation of Text, Audio, Image, and Video

Alexander Htet Kyaw, Lenin Ravindranath Sivalingam

机构 * Massachusetts Institute of Technology（麻省理工学院） ； Microsoft Research（微软研究院） 纠错

Comments Accepted to NeurIPS 2025, Conference on Neural Information Processing Systems, Workshop on Generative and Protective AI for Content Creation

A Principle of Targeted Intervention for Multi-Agent Reinforcement Learning

Anjie Liu, Jianhong Wang, Samuel Kaski, Jun Wang, Mengyue Yang

机构 * HKUST (GZ)（香港科技大学（广州）） ； INFORMED-AI Hub（INFORMED-AI中心） ； University of Bristol（布里斯托大学） ； ELLIS Institute Finland（芬兰ELLIS研究所） ； Centre for Artificial Intelligence（人工智能中心） ； University College London（伦敦大学学院） ； School of Engineering Mathematics and Technology（工程数学与技术学院） 纠错

Comments Published in NeurIPS 2025

HiMaCon: Discovering Hierarchical Manipulation Concepts from Unlabeled Multi-Modal Data

Ruizhe Liu, Pei Zhou, Qian Luo, Li Sun, Jun Cen, Yibing Song, Yanchao Yang

机构 * HKU Musketeers Foundation Institute of Data Science（香港大学数据科学学院） ； The University of Hong Kong（香港大学） ； DAMO Academy（达摩院） ； Alibaba Group（阿里巴巴集团） 纠错

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

FlexAC: Towards Flexible Control of Associative Reasoning in Multimodal Large Language Models

Shengming Yuan, Xinyu Lyu, Shuailong Wang, Beitao Chen, Jingkuan Song, Lianli Gao

机构 * University of Electronic Science and Technology of China（电子科学与技术大学） ； Southwestern University of Finance and Economics（西南财经大学） ； Tongji University（同济大学） 纠错

Comments 19 pages, 11 figures. Accepted by the 39th Conference on Neural Information Processing Systems (NeurIPS 2025)

Deep Graph Learning for Industrial Carbon Emission Analysis and Policy Impact

Xuanming Zhang

机构 * Stanford University（斯坦福大学） 纠错

Comments NeurIPS 2025 AI for Science Workshop

AutoVLA: A Vision-Language-Action Model for End-to-End Autonomous Driving with Adaptive Reasoning and Reinforcement Fine-Tuning

Zewei Zhou, Tianhui Cai, Seth Z. Zhao, Yun Zhang, Zhiyu Huang, Bolei Zhou, Jiaqi Ma

机构 * University of California, Los Angeles（加州大学洛杉矶分校） 纠错

Comments NeurIPS 2025; Website link:https://autovla.github.io/

Reasoning Models Hallucinate More: Factuality-Aware Reinforcement Learning for Large Reasoning Models

Junyi Li, Hwee Tou Ng

机构 * Department of Computer Science, National University of Singapore（计算机科学系，新加坡国立大学） 纠错

Comments accepted by NeurIPS 2025

Critical Batch Size Revisited: A Simple Empirical Approach to Large-Batch Language Model Training

William Merrill, Shane Arora, Dirk Groeneveld, Hannaneh Hajishirzi

机构 * Allen Institute for AI（人工智能研究所） 纠错

Comments Neurips 2025

Mustafar: Promoting Unstructured Sparsity for KV Cache Pruning in LLM Inference

Donghyeon Joo, Helya Hosseini, Ramyad Hadidi, Bahar Asgari

机构 * Department of Computer Science, University of Maryland（计算机科学系，马里兰大学） ； d-Matrix 纠错

Comments 20 pages, 9 figures, NeurIPS 2025

Two Causally Related Needles in a Video Haystack

Miaoyu Li, Qin Chao, Boyang Li

机构 * College of Computing and Data Science（计算与数据科学学院） ； Nanyang Technological University（南洋理工大学） 纠错

Comments Accepted to NeurIPS 2025 D&B Track

Exact Expressive Power of Transformers with Padding

William Merrill, Ashish Sabharwal

机构 * Allen Institute for AI（艾伦人工智能研究所） 纠错

Comments Neurips 2025

DOVE: Efficient One-Step Diffusion Model for Real-World Video Super-Resolution

Zheng Chen, Zichen Zou, Kewei Zhang, Xiongfei Su, Xin Yuan, Yong Guo, Yulun Zhang

机构 * School of Computer Science, Shanghai Jiao Tong University（上海交通大学计算机科学学院） ； Zhiyuan College, Shanghai Jiao Tong University（上海交通大学紫阳学院） ； China Mobile Research Institute（中国移动研究院） ； Westlake University（西湖大学） ； Huawei Consumer Business Group（华为消费者业务集团） 纠错

Comments Accepted to NeurIPS 2025. Code is available at: https://github.com/zhengchen1999/DOVE

Learning Dynamics of RNNs in Closed-Loop Environments

Yoav Ger, Omri Barak

机构 * Rappaport Faculty of Medicine and Network Biology Research Laboratory（拉普波特医学与网络生物学研究实验室） ； Technion, Israel Institute of Technology（技术学院，以色列理工学院） 纠错

Comments Accepted at NeurIPS 2025

RadZero: Similarity-Based Cross-Attention for Explainable Vision-Language Alignment in Chest X-ray with Zero-Shot Multi-Task Capability

Jonggwon Park, Byungmu Yoon, Soobum Kim, Kyoyun Choi

Comments NeurIPS 2025

A Little Depth Goes a Long Way: The Expressive Power of Log-Depth Transformers

William Merrill, Ashish Sabharwal

机构 * Allen Institute for AI（阿伦人工智能研究所） ； New York University（纽约大学） 纠错

Comments NeurIPS 2025

SORTeD Rashomon Sets of Sparse Decision Trees: Anytime Enumeration

Elif Arslan, Jacobus G. M. van der Linden, Serge Hoogendoorn, Marco Rinaldi, Emir Demirović

机构 * Delft University of Technology（代尔夫特理工大学） 纠错

Comments 32 pages, 10 figures, to be published in the proceedings of The Thirty-Ninth Annual Conference on Neural Information Processing Systems

A Probabilistic Approach to Pose Synchronization for Multi-Reference Alignment with Applications to MIMO Wireless Communication Systems

Rob Romijnders, Gabriele Cesa, Christos Louizos, Kumar Pratik, Arash Behboodi

机构 * University of Amsterdam（阿姆斯特丹大学） ； QUvA-Lab（QUvA实验室） ； Qualcomm AI Research（高通人工智能研究） 纠错

Comments To appear in NeurIPS workshop: AI and ML for Next-Generation Wireless Communications (AI4NextG)