Continuous Thought Machines
机构 * Sakana AI ； University of Tsukuba（东京大学） ； IT University of Copenhagen（哥本哈根IT大学）
Comments Technical report accompanied by online project page: https://pub.sakana.ai/ctm/ Accepted as a spotlight paper at NeurIPS 2025
期刊&会议
NeurIPS
Conference on Neural Information Processing Systems · 会议 · Machine Learning
Toward a Holistic Evaluation of Robustness in CLIP Models
Comments Accepted to IEEE TPAMI, extension of NeurIPS'23 work: A Closer Look at the Robustness of Contrastive Language-Image Pre-Training (CLIP)
Knowledge Distillation Detection for Open-weights Models
机构 * Department of Statistics, Purdue University（普渡大学统计学系） ； Department of Computer Science, Purdue University（普渡大学计算机科学系）
Comments NeurIPS 2025
Diffusion Transformers for Imputation: Statistical Efficiency and Uncertainty Quantification
Comments 49 pages, 4 figures. Accepted as a poster at NeurIPS 2025
Catalyst GFlowNet for electrocatalyst design: A hydrogen evolution reaction case study
机构 * University of Waterloo（滑铁卢大学） ； ETH Zurich（苏黎世联邦理工学院） ； Mila Quebec AI Institute（魁北克AI研究所）
Comments 5 pages, 2 figures. Accepted to NeurIPS AI for Materials Workshop 2025
FairContrast: Enhancing Fairness through Contrastive learning and Customized Augmenting Methods on Tabular Data
机构 * Department of Industrial Engineering University of Central Florida（工业工程系，中央佛罗里达大学） ； Department of Computer Science University of Central Florida（计算机科学系，中央佛罗里达大学）
Comments Accepted to NeurIPS 2025 - Reliable ML Workshop
MPMAvatar: Learning 3D Gaussian Avatars with Accurate and Robust Physics-Based Dynamics
机构 * KAIST（韩国科学技术院）
Comments Accepted to NeurIPS 2025
Pharmacophore-Guided Generative Design of Novel Drug-Like Molecules
机构 * Center for AI in Chemistry, ITMO University（人工智能化学中心，ITMO大学） ； Moscow Center for Advanced Studies（莫斯科高级研究所在） ； D ONE AG（D ONE公司）
Comments AI4Mat-NeurIPS-2025 Poster
MEMTRACK: Evaluating Long-Term Memory and State Tracking in Multi-Platform Dynamic Agent Environments
Comments Accepted to NeurIPS 2025 SEA Workshop
The Social Laboratory: A Psychometric Framework for Multi-Agent LLM Evaluation
机构 * Independent researcher（独立研究者）
Comments 39th Conference on Neural Information Processing Systems (NeurIPS 2025) Workshop on Evaluating the Evolving LLM Lifecycle: Benchmarks, Emergent Abilities, and Scaling
SSTAG: Structure-Aware Self-Supervised Learning Method for Text-Attributed Graphs
机构 * Institute of Information Engineering, CAS（信息工程研究所，中国科学院） ； School of Cyberspace Security, UCAS（网络安全学院，中国科学院大学） ； Wuhan University of Technology（武汉理工大学） ； Xiaomi EV（小米汽车） ； Renmin University of China（中国人民大学）
Comments Accepted by NeurIPS 2025
Detoxifying Large Language Models via Autoregressive Reward Guided Representation Editing
机构 * SKLCCSE, Beihang University（北航智能计算与复杂系统研究院） ； National University of Singapore（新加坡国立大学） ； Zhongguancun Laboratory, Beijing（中关村实验室） ； Institute of Dataspace, Hefei（合肥数据空间研究院） ； Nanyang Technological University（南洋理工大学）
Comments Accepted to NeurIPS 25
Evaluating LLMs for Combinatorial Optimization: One-Phase and Two-Phase Heuristics for 2D Bin-Packing
机构 * Department of Computer Science, School of Science and Technology（计算机科学系，科学与技术学院） ； University of London（伦敦大学） ； BRAC University（BRAC大学）
Comments 1 table, 6 figures. 39th Conference on Neural Information Processing Systems (NeurIPS 2025) Accepted for the Workshop: Evaluating the Evolving LLM Lifecycle Benchmarks, Emergent Abilities, and Scaling
CAMILA: Context-Aware Masking for Image Editing with Language Alignment
Comments Accepted by NeurIPS 2025
Semantic and Visual Crop-Guided Diffusion Models for Heterogeneous Tissue Synthesis in Histopathology
Comments NeurIPS 2025
Autoregressive Adversarial Post-Training for Real-Time Interactive Video Generation
Comments NeurIPS 2025
Dynamic Bundling with Large Language Models for Zero-Shot Inference on Text-Attributed Graphs
机构 * Peking University（北京大学） ； Nanyang Technological University（南洋理工大学） ； University of Washington（华盛顿大学） ； University of California, Los Angeles（加州大学洛杉矶分校） ； University of Illinois Chicago（伊利诺伊大学芝加哥分校）
Comments Accepted by NeurIPS 2025
AgentDAM: Privacy Leakage Evaluation for Autonomous Web Agents
机构 * FAIR at Meta（Meta 的公平性研究部门）
Comments Accepted to NeurIPS 2025 (D&B track), project page: https://github.com/facebookresearch/ai-agent-privacy
Mol-LLaMA: Towards General Understanding of Molecules in Large Molecular Language Model
机构 * KAIST（韩国科学技术院）
Comments NeurIPS 2025
Diffusion Model as a Noise-Aware Latent Reward Model for Step-Level Preference Optimization
机构 * MAIS, CASIA（中国科学院自动化研究所MAIS实验室） ； Kuaishou Technology（快手科技） ； School of Artificial Intelligence, UCAS（中国科学院大学人工智能学院）
Comments NeurIPS 2025
GlobalTomo: A global dataset for physics-ML seismic wavefield modeling and FWI
Comments 50 pages, NeurIPS DB 2025
DiffCut: Catalyzing Zero-Shot Semantic Segmentation with Diffusion Features and Recursive Normalized Cut
Comments NeurIPS 2024. Project page at https://diffcut-segmentation.github.io. Code at https://github.com/PaulCouairon/DiffCut
Instant4D: 4D Gaussian Splatting in Minutes
机构 * University of Pittsburgh（匹兹堡大学） ； Carnegie Mellon University（卡内基梅隆大学）
Comments Accepted by NeurIPS 25
Topology of Reasoning: Understanding Large Reasoning Models through Reasoning Graph Properties
机构 * The University of Tokyo（东京大学）
Comments Accepted to Neurips 2025
Prompt Tuning Decision Transformers with Structured and Scalable Bandits
机构 * Örebro University（奥雷布罗大学） ； King AI Labs, Microsoft Gaming（King AI实验室，微软游戏）
Comments Accepted at NeurIPS 2025
Benchmarking Agentic Systems in Automated Scientific Information Extraction with ChemX
机构 * Center for AI in Chemistry, ITMO University（人工智能化学中心，ITMO大学） ； D ONE AG（D ONE公司）
Comments Accepted at The AI for Accelerated Materials Discovery (AI4Mat) Workshop, NeurIPS 2025
Efficient Multi-modal Large Language Models via Progressive Consistency Distillation
Comments Accepted by NeurIPS 2025
Looking Beyond the Known: Towards a Data Discovery Guided Open-World Object Detection
机构 * The University of Texas at Dallas（德克萨斯大学达拉斯分校）
Comments Accepted to NeurIPS'25. 22 pages, 6 figures
Beyond Token Probes: Hallucination Detection via Activation Tensors with ACT-ViT
机构 * Technion（技术离子大学） ； University of Groningen（格罗宁根大学） ； Nvidia Research（英伟达研究）
Comments Published in NeurIPS 2025
NeurIPS should lead scientific consensus on AI policy
机构 * Stanford Institute for Human-Centered Artificial Intelligence (HAI)（斯坦福大学人本人工智能研究所）
Comments Published at NeurIPS 2025
