NeurIPS

Uncertainty Quantification for Reduced-Order Surrogate Models Applied to Cloud Microphysics

Jonas E. Katona, Emily K. de Jong, Nipun Gunawardena

机构 * Department of Applied & Computational Mathematics（应用与计算数学系） ； Yale University（耶鲁大学） ； Atmospheric, Earth, & Energy Division（大气、地球与能源部门） ； Lawrence Livermore National Laboratory（劳伦斯利弗莫尔国家实验室） 纠错

Comments Accepted at the NeurIPS 2025 Workshop on Machine Learning and the Physical Sciences (ML4PS). 12 pages, 3 figures, 2 tables. LLNL-CONF-2010541

Learning at the Speed of Physics: Equilibrium Propagation on Oscillator Ising Machines

Alex Gower

机构 * University of Cambridge（剑桥大学） ； Nokia Bell Labs（诺基亚贝尔实验室） 纠错

Comments 5 pages, 2 figures, NeurIPS 2025 Machine Learning and the Physical Sciences (ML4PS)

What One Cannot, Two Can: Two-Layer Transformers Provably Represent Induction Heads on Any-Order Markov Chains

Chanakya Ekbote, Marco Bondaschi, Nived Rajaraman, Jason D. Lee, Michael Gastpar, Ashok Vardhan Makkuva, Paul Pu Liang

机构 * Massachusetts Institute of Technology（麻省理工学院） ； École Polytechnique Fédérale de Lausanne（瑞士联邦理工学院） ； University of California, Berkeley（加州大学伯克利分校） ； Princeton University（普林斯顿大学） 纠错

Comments Accepted at NeurIPS 2025

VITRIX-CLIPIN: Enhancing Fine-Grained Visual Understanding in CLIP via Instruction Editing Data and Long Captions

Ziteng Wang, Siqi Yang, Limeng Qiao, Lin Ma

Comments Accepted to NeurIPS 2025

Private Evolution Converges

Tomás González, Giulia Fanti, Aaditya Ramdas

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

Comments Accepted for NeurIPS 2025

TAPIP3D: Tracking Any Point in Persistent 3D Geometry

Bowei Zhang, Lei Ke, Adam W. Harley, Katerina Fragkiadaki

机构 * Carnegie Mellon University（卡内基梅隆大学） ； Peking University（北京大学） ； Stanford University（斯坦福大学） 纠错

Comments NeurIPS 2025. Long-term feed-forward 3D point tracking in persistent 3D point maps. Code:https://github.com/zbw001/TAPIP3D

Harnessing the Computation Redundancy in ViTs to Boost Adversarial Transferability

Jiani Liu, Zhiyuan Wang, Zeliang Zhang, Chao Huang, Susan Liang, Yunlong Tang, Chenliang Xu

机构 * University of Rochester（罗切斯特大学） ； University of California, Santa Barbara（加州大学圣巴巴拉分校） 纠错

Comments Accepted by NeurIPS 2025

Open-Insect: Benchmarking Open-Set Recognition of Novel Species in Biodiversity Monitoring

Yuyan Chen, Nico Lang, B. Christian Schmidt, Aditya Jain, Yves Basset, Sara Beery, Maxim Larrivée, David Rolnick

机构 * McGill University（麦吉尔大学） ； Mila - Quebec Artificial Intelligence Institute（魁北克人工智能研究所） ； University of Copenhagen（哥本哈根大学） ； Agriculture and Agri-food Canada（加拿大农业与农业食品部） ； Smithsonian Tropical Research Institute（史密松尼热带研究所） ； Biology Center, Czech Academy of Sciences（捷克科学院生物学中心） ； Maestria de Entomologia, University of Panama（巴拿马大学昆虫学系） ； Massachusetts Institute of Technology（麻省理工学院） ； Montréal Insectarium（蒙特利尔昆虫馆） 纠错

Comments NeurIPS 2025 Dataset and Benchmark Track (Spotlight); Code and data are available at https://yuyan-c.github.io/open-insect-project/

Simultaneous Swap Regret Minimization via KL-Calibration

Haipeng Luo, Spandan Senapati, Vatsal Sharan

机构 * USC（美国大学） 纠错

Comments NeurIPS 2025 (Spotlight)

Intrinsic Dimension Estimation for Radio Galaxy Zoo using Diffusion Models

Joan Font-Quer Roset, Devina Mohan, Anna Scaife

机构 * Department of Computer Science（计算机科学系） ； The University of Manchester（曼彻斯特大学） ； Department of Physics and Astronomy（物理与天文学系） ； The Alan Turing Institute（艾伦·图灵研究所） 纠错

Comments 9 pages, 5 figures, 2 tables, submitted to NeurIPS 2025 ML4PS Workshop

Journal ref NeurIPS 2025 Workshop on ML for Physical Sciences

Power law attention biases for molecular transformers

Jay Shen, Yifeng Tang, Andrew Ferguson

Comments Machine Learning and the Physical Sciences Workshop at NeurIPS 2025

Data-efficient U-Net for Segmentation of Carbide Microstructures in SEM Images of Steel Alloys

Alinda Ezgi Gerçek, Till Korten, Paul Chekhonin, Maleeha Hassan, Peter Steinbach

机构 * Helmholtz AI Team Matter (FWCC-A)（海德堡人工智能团队物质（FWCC-A）） ； Helmholtz-Zentrum Dresden-Rossendorf（德罗斯登-罗斯托克研究中心） ； Institute of Resource Ecology, Structural Materials Department（资源生态研究所，结构材料部门） 纠错

Journal ref Machine Learning and the Physical Sciences Workshop @ NeurIPS 2025 https://openreview.net/forum?id=xYY5pn4f8N

Sparse Methods for Vector Embeddings of TPC Data

Tyler Wheeler, Michelle P. Kuchera, Raghuram Ramanujan, Ryan Krupp, Chris Wrede, Saiprasad Ravishankar, Connor L. Cross, Hoi Yan Ian Heung, Andrew J. Jones, Benjamin Votaw

机构 * Michigan State University（密歇根州立大学） ； Davidson College（戴维森学院） 纠错

Comments NeurIPS Machine Learning and the Physical Sciences Workshop 2025

Power Ensemble Aggregation for Improved Extreme Event AI Prediction

Julien Collard, Pierre Gentine, Tian Zheng

机构 * Ecole Polytechnique（巴黎政治学院） ； LEAP, Columbia University（哥伦比亚大学LEAP） ； Department of Statistics, Columbia University（哥伦比亚大学统计系） 纠错

Comments Accepted for the NeurIPS 2025 ML4PS workshop

One-Shot Transfer Learning for Nonlinear PDEs with Perturbative PINNs

Samuel Auroy, Pavlos Protopapas

机构 * ENSTA Paris（巴黎国立应用科学大学） ； Harvard University（哈佛大学） 纠错

Comments Accepted at Machine Learning and the Physical Sciences Workshop, NeurIPS 2025

Sheaf Cohomology of Linear Predictive Coding Networks

Jeffrey Seely

机构 * Sakana AI 纠错

Comments Accepted to NeurIPS 2025 Workshop on Symmetry and Geometry in Neural Representations

Sequentially Auditing Differential Privacy

Tomás González, Mateo Dulce-Rubio, Aaditya Ramdas, Mónica Ribero

机构 * Carnegie Mellon University（卡内基梅隆大学） ； Google Research（谷歌研究） 纠错

Comments Accepted in NeurIPS 2025

Neuro-Spectral Architectures for Causal Physics-Informed Networks

Arthur Bizzi, Leonardo M. Moreira, Márcio Marques, Leonardo Mendonça, Christian Júnior de Oliveira, Vitor Balestro, Lucas dos Santos Fernandez, Daniel Yukimura, Pavel Petrov, João M. Pereira, Tiago Novello, Lucas Nissenbaum

机构 * École Polytechnique Fédérale de Lausanne (EPFL)（联邦理工学院洛桑校区） ； Instituto de Matemática Pura e Aplicada (IMPA)（数学与应用数学研究所） ； Universidade do Estado do Rio de Janeiro (UERJ)（里约热内卢州立大学） ； Laboratório Nacional de Computação Científica (LNCC)（国家科学计算实验室） ； University of Georgia (UGA)（佐治亚大学） 纠错

Comments Accepted at NeurIPS 2025 (poster). 24 pages, 10 figures

Preserving Task-Relevant Information Under Linear Concept Removal

Floris Holstege, Shauli Ravfogel, Bram Wouters

机构 * University of Amsterdam, Department of Quantitative Economics（阿姆斯特丹大学量化经济学系） ； New York University, Center for Data Science（纽约大学数据科学中心） ； Tinbergen Institute（廷伯根研究所） 纠错

Comments Published at NeurIPS 2025

BecomingLit: Relightable Gaussian Avatars with Hybrid Neural Shading

Jonathan Schmidt, Simon Giebenhain, Matthias Niessner

机构 * Technical University of Munich（慕尼黑技术大学） 纠错

Comments NeurIPS 2025, Project Page: see https://jonathsch.github.io/becominglit/ , YouTube Video: see https://youtu.be/xPyeIqKdszA

Latent Principle Discovery for Language Model Self-Improvement

Keshav Ramji, Tahira Naseem, Ramón Fernandez Astudillo

机构 * IBM Research AI（IBM人工智能研究） 纠错

Comments Accepted at NeurIPS 2025

Nonlinear Laplacians: Tunable principal component analysis under directional prior information

Yuxin Ma, Dmitriy Kunisky

Comments 54 pages, 6 figures, closest to version to be published in NeurIPS 2025

Beyond $\tilde{O}(\sqrt{T})$ Constraint Violation for Online Convex Optimization with Adversarial Constraints

Abhishek Sinha, Rahul Vaze

Comments To appear in NeurIPS 2025

The Empty Chair: Using LLMs to Raise Missing Perspectives in Policy Deliberations

Suyash Fulay, Dimitra Dimitrakopoulou, Deb Roy

机构 * Center for Constructive Communication（建设性沟通中心） ； MIT（麻省理工学院） 纠错

Comments 39th Conference on Neural Information Processing Systems (NeurIPS 2025) Workshop: PersonaLLM: Workshop on LLM Persona Modeling

Enhancing Demand-Oriented Regionalization with Agentic AI and Local Heterogeneous Data for Adaptation Planning

Seyedeh Mobina Noorani, Shangde Gao, Changjie Chen, Karla Saldana Ochoa

机构 * College of Design, Construction and Planning, University of Florida（佛罗里达大学设计、建设与规划学院） ； Department of Electrical and Computer Engineering, University of Florida（佛罗里达大学电气与计算机工程系） ； Florida Institute for Built Environment Resilience (FIBER), University of Florida（佛罗里达大学环境韧性研究所） 纠错

Comments Accepted by NeurIPS 2025 UrbanAI Workshop as poster

Towards Universal Neural Operators through Multiphysics Pretraining

Mikhail Masliaev, Dmitry Gusarov, Ilya Markov, Alexander Hvatov

机构 * ITMO University（ITMO大学） 纠错

Comments 5 pages, 1 figure, accepted for Machine Learning and the Physical Sciences Workshop, NeurIPS 2025

Faithful Summarization of Consumer Health Queries: A Cross-Lingual Framework with LLMs

Ajwad Abrar, Nafisa Tabassum Oeshy, Prianka Maheru, Farzana Tabassum, Tareque Mohmud Chowdhury

机构 * Islamic University of Technology（伊斯兰技术大学） 纠错

Comments Accepted at the 5th Muslims in Machine Learning (MusIML) Workshop, co-located with NeurIPS 2025

Fast Data Attribution for Text-to-Image Models

Sheng-Yu Wang, Aaron Hertzmann, Alexei A Efros, Richard Zhang, Jun-Yan Zhu

机构 * Carnegie Mellon University（卡内基梅隆大学） ； Adobe Research（Adobe研究） ； UC Berkeley（伯克利大学） 纠错

Comments NeurIPS 2025 camera ready. Project page: https://peterwang512.github.io/FastGDA

Saying the Unsaid: Revealing the Hidden Language of Multimodal Systems Through Telephone Games

Juntu Zhao, Jialing Zhang, Chongxuan Li, Dequan Wang

机构 * Shanghai Jiao Tong University（上海交通大学） ； Renmin University of China（中国人民大学） 纠错

Comments Accepted by NeurIPS 2025 MTI-LLM Workshop

Bayesian Evaluation of Large Language Model Behavior

Rachel Longjohn, Shang Wu, Saatvik Kher, Catarina Belém, Padhraic Smyth

机构 * Department of Statistics, University of California, Irvine（统计系，加州大学伊文斯分校） ； Department of Computer Science, University of California, Irvine（计算机科学系，加州大学伊文斯分校） 纠错

Comments Accepted to NeurIPS 2025 Workshop on Evaluating the Evolving LLM Lifecycle: Benchmarks, Emergent Abilities, and Scaling