Self-Improving Embodied Foundation Models
机构 * Generalist ； Google DeepMind（谷歌DeepMind）
Comments Appearing in the Conference on Neural Information Processing Systems (NeurIPS 2025)
期刊&会议
NeurIPS
Conference on Neural Information Processing Systems · 会议 · Machine Learning
OmniSegmentor: A Flexible Multi-Modal Learning Framework for Semantic Segmentation
机构 * VCIP, CS, Nankai University（视觉计算研究所，计算机科学，南开大学）
Comments Accepted by NeurIPS 2025
Moment- and Power-Spectrum-Based Gaussianity Regularization for Text-to-Image Models
机构 * KAIST（韩国科学技术院）
Comments Accepted to NeurIPS 2025
ThinkAct: Vision-Language-Action Reasoning via Reinforced Visual Latent Planning
机构 * NVIDIA ； National Taiwan University（国立台湾大学）
Comments NeurIPS 2025. Project page: https://jasper0314-huang.github.io/thinkact-vla/
ForceVLA: Enhancing VLA Models with a Force-aware MoE for Contact-rich Manipulation
机构 * Fudan University（复旦大学） ； Shanghai Jiao Tong University（上海交通大学） ； Shanghai Innovation Institute（上海创新研究院） ； Shanghai AI Lab（上海人工智能实验室） ； National University of Singapore（新加坡国立大学） ； Shanghai University（上海大学） ； Xi’an Jiaotong University（西安交通大学） ； Noematrix Intelligence（Noematrix智能科技）
Comments NeurIPS 2025
OmniSync: Towards Universal Lip Synchronization via Diffusion Transformers
机构 * Renmin University of China（中国人民大学） ； Kling Team, Kuaishou Technology（快手科技 Kling 团队） ； Tsinghua University（清华大学）
Comments Accepted as NeurIPS 2025 spotlight
SWAT: Sliding Window Adversarial Training for Gradual Domain Adaptation
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 电子科学与技术大学）
Comments submitted to NIPS 2025
Learning Environment-Aware Affordance for 3D Articulated Object Manipulation under Occlusions
机构 * CFCS, School of CS, PKU（计算机学院，北京大学） ； School of EECS, PKU（电子工程系，北京大学） ； University of Oxford（牛津大学） ； National Key Laboratory for Multimedia Information Processing, School of CS, PKU（多媒体信息处理国家重点实验室，计算机学院，北京大学）
Comments In 37th Conference on Neural Information Processing Systems (NeurIPS 2023). Website at https://chengkaiacademycity.github.io/EnvAwareAfford/
Quantum Noise Tomography with Physics-Informed Neural Networks
机构 * Lawrence Berkeley National Lab（伯克利国家实验室）
Comments 6 pages, 3 figures, Machine Learning and the Physical Sciences Workshop at the 39th conference on Neural Information Processing Systems (NeurIPS)
Multilingual Diversity Improves Vision-Language Representations
机构 * University of Washington（华盛顿大学） ； Allen Institute for Artificial Intelligence（人工智能研究院）
Comments NeurIPS 2024 Spotlight paper
A Permutation-free Kernel Two-Sample Test
Comments Published at the Thirty-sixth Conference on Neural Information Processing Systems (NeurIPS), with an oral presentation. The current version on arXiv fixes a bug in the proof of Theorem 9
The LLM as a Network Operator: A Vision for Generative AI in the 6G Radio Access Network
机构 * African Institute for Mathematical Sciences（非洲数学科学研究所）
Comments Submitted to Workshop on AI and ML for Next-Generation Wireless Communications and Networking, NeurIPS 2025
GeoDE: a Geographically Diverse Evaluation Dataset for Object Recognition
机构 * Princeton University（普林斯顿大学） ； Stanford University（斯坦福大学） ； Meta AI ； Runway
Comments Published at NeurIPS D&B, 2023
A Logic for Expressing Log-Precision Transformers
机构 * New York University（纽约大学） ； Allen Institute for AI（人工智能研究院）
Comments May 24, 2023: Restructured version of old preprint. Oct 12, 2023: To appear at NeurIPS. Sept 10, 2025: minor technical corrections
An Inexact Proximal Framework for Nonsmooth Riemannian Difference-of-Convex Optimization
Comments A preliminary version was submitted to NeurIPS 2024
Exploratory Retrieval-Augmented Planning For Continual Embodied Instruction Following
机构 * Department of Computer Science and Engineering, Sungkyunkwan University（全南大学计算机科学与工程系） ； Acryl Inc.（阿克罗尔公司）
Comments 21 pages. NeurIPS 2024
Journal ref Advances in Neural Information Processing Systems 37, 67034-67060, 2024
Interpretable Image Classification with Adaptive Prototype-based Vision Transformers
机构 * Dartmouth College（达特茅斯学院） ； Duke University（杜克大学） ； University of Maine（缅因大学）
Journal ref Neural Information Processing Systems, 2024
Using topological data analysis to compare inter-subject variability across resting state functional MRI brain representations
Comments first version submitted for NeurIPS 2023 conference; second version revised and submitted to SOCG 2024; current version submitted to the journal "Imaging Neuroscience."
On the Benefits of Public Representations for Private Transfer Learning under Distribution Shift
机构 * Carnegie Mellon University（卡内基梅隆大学）
Comments Published in NeurIPS 2024
Efficient Methods for Non-stationary Online Learning
机构 * National Key Laboratory for Novel Software Technology, Nanjing University, China（国家新型软件技术重点实验室，南京大学） ； School of Artificial Intelligence, Nanjing University, China（人工智能学院，南京大学）
Comments V3 changes: accepted by JMLR 2025 and improve the writing; V2/V1 changes: investigate interval dynamic regret and add two applications (online non-stochastic control and online PCA) and improve the presentation; preliminary version published at NeurIPS'22
Journal ref Journal of Machine Learning Research, 2025
NeuroBOLT: Resting-state EEG-to-fMRI Synthesis with Multi-dimensional Feature Mapping
机构 * Vanderbilt University（范德比尔特大学） ； Vanderbilt University Medical Center（范德比尔特大学医学中心）
Comments This preprint has been accepted to NeurIPS 2024
The Curious Price of Distributional Robustness in Reinforcement Learning with a Generative Model
机构 * JHU Department of Electrical and Computer Engineering, Johns Hopkins University, MD, USA（约翰霍普金斯大学电气与计算机工程系） ； CUHK Department of Statistics, The Chinese University of Hong Kong, Hong Kong（香港中文大学统计系） ； UPenn Department of Statistics and Data Science, Wharton School, University of Pennsylvania, Philadelphia, PA 19104, USA（宾夕法尼亚大学统计与数据科学系） ； ESP Earth Species Project（地球物种计划） ； Yale Department of Statistics and Data Science, Yale University, CT, USA（耶鲁大学统计与数据科学系）
Comments A short version was published in Neural Information Processing Systems (2023); Under Submission
Unmasking COVID-19 Vulnerability in Nigeria: Mapping Risks Beyond Urban Hotspots
机构 * Carnegie Mellon University（卡内基梅隆大学）
Comments 8 pages, 6 figures. Submission to NeurIPS 2025 in preparation
Slice-100K: A Multimodal Dataset for Extrusion-based 3D Printing
机构 * Iowa State University（爱荷华州立大学） ； New York University（纽约大学）
Comments Accepted to NeurIPS 2024. For codebase, see https://github.com/idealab-isu/Slice-100K
PixFoundation 2.0: Do Video Multi-Modal LLMs Use Motion in Visual Grounding?
Comments Work under review in NeurIPS 2025 with the title "Are we using Motion in Referring Segmentation? A Motion-Centric Evaluation"
Towards Agentic AI on Particle Accelerators
机构 * Helmholtz Zentrum Berlin（柏林海德堡中心） ； LBNL（劳伦斯伯克利国家实验室） ； DESY（德意志电子同步辐射研究中心） ； FNAL（费米国家加速器实验室）
Comments 5 pages, 3 figures, Machine Learning and the Physical Sciences at Workshop at the 38th conference on Neural Information Processing Systems (NeurIPS)
Journal ref Machine Learning and the Physical Sciences Workshop at the 38th conference on Neural Information Processing Systems (NeurIPS) December 15, 2024
Analysis of Error Sources in LLM-based Hypothesis Search for Few-Shot Rule Induction
机构 * Department of Statistics, University of California, Los Angeles（统计学系，加州大学洛杉矶分校） ； Microsoft, Redmond, WA（微软公司，西雅图，华盛顿州） ； Department of Psychology, University of California, Los Angeles（心理学系，加州大学洛杉矶分校）
Comments This is the preprint version corresponding to our NeurIPS 2025 Workshop on Multimodal Algorithmic Reasoning submission
First-Place Solution to NeurIPS 2024 Invisible Watermark Removal Challenge
机构 * Mohamed bin Zayed University of Artificial Intelligence (MBZUAI)（Mohamed bin Zayed大学人工智能学院） ； Michigan State University (MSU)（密歇根州立大学）
Comments Winning solution to the NeurIPS 2024 Erasing the Invisible challenge
Real-Time Intuitive AI Drawing System for Collaboration: Enhancing Human Creativity through Formal and Contextual Intent Integration
机构 * SNU（首尔国立大学）
Comments 6 pages, 4 figures, NeurIPS Creative AI Track 2025
A New NMT Model for Translating Clinical Texts from English to Spanish
机构 * College of Information and Computer Science, University of Massachusetts Amherst（马萨诸塞大学阿姆赫斯特信息与计算机科学学院） ； Department of Computer Sciences, University of Massachusetts Lowell（马萨诸塞大学洛约拉计算机科学系） ； Department of Quantitative Health Sciences, University of Massachusetts Medical School（马萨诸塞大学医学学院定量健康科学系） ； Bedford VAMC and CHOIR（贝德福德退伍军人医疗中心和CHOIR）
Comments This work was accepted by the Machine Learning for Health (ML4H) Workshop at NeurIPS 2018
