Explanation-Driven Counterfactual Testing for Faithfulness in Vision-Language Model Explanations
机构 * Mercedes-Benz Research & Development North America（梅赛德斯-奔驰研究与开发北美）
Comments NeurIPS 2025 workshop on Regulatable ML
期刊&会议
NeurIPS
Conference on Neural Information Processing Systems · 会议 · Machine Learning
Steering When Necessary: Flexible Steering Large Language Models with Backtracking
机构 * State Key Laboratory for Novel Software Technology, Nanjing University, China（新型软件技术国家重点实验室，南京大学，中国）
Comments NeurIPS 2025
Post Hoc Regression Refinement via Pairwise Rankings
机构 * MPI-INF（马克斯·普朗克研究所信息部门） ； MIT（麻省理工学院）
Comments NeurIPS 2025 camera-ready version
Robust Estimation Under Heterogeneous Corruption Rates
机构 * University of California, Berkeley（加州大学伯克利分校） ； University of Washington（华盛顿大学）
Comments NeurIPS 2025, fixed PAC minimax definition
Precise Information Control in Long-Form Text Generation
机构 * Paul G. Allen School of Computing Science & Engineering, University of Washington（保罗·G·艾伦计算科学与工程学院，华盛顿大学） ； Princeton Language and Intelligence (PLI), Princeton University（普林斯顿语言与智能（PLI），普林斯顿大学） ； Allen Institute for AI（艾伦人工智能研究所）
Comments NeurIPS 2025. 66 pages, 8 figures. Code is at https://github.com/jacqueline-he/precise-information-control
ATLAS: Autoformalizing Theorems through Lifting, Augmentation, and Synthesis of Data
Comments Accepted to NeurIPS 2025
A distributional simplicity bias in the learning dynamics of transformers
机构 * International School for Advanced Studies（国际先进研究学校）
Comments 10 pages, 5 figures, NeurIPS 2024
Journal ref NeurIPS 2024
Stable Cinemetrics : Structured Taxonomy and Evaluation for Professional Video Generation
机构 * Stability AI ； Arizona State University（亚利桑那州立大学） ； Google DeepMind（谷歌DeepMind）
Comments NeurIPS 2025. Project Page : https://stable-cinemetrics.github.io/
Deep Taxonomic Networks for Unsupervised Hierarchical Prototype Discovery
机构 * Georgia Institute of Technology（佐治亚理工学院） ； University of Virginia（弗吉尼亚大学）
Comments NeurIPS 2025
Journal ref NeurIPS 2025
MaskSQL: Safeguarding Privacy for LLM-Based Text-to-SQL via Abstraction
机构 * University of Waterloo（多伦多大学） ； Vector Institute（向量研究所） ； Layer 6 AI
Comments Accepted to the 3rd Workshop on Regulatable ML at NeurIPS 2025
Scalable Fingerprinting of Large Language Models
机构 * University of Washington（华盛顿大学） ； Sentient
Comments Spotlight at NeurIPS 2025
Unlocking Transfer Learning for Open-World Few-Shot Recognition
Comments Accepted at NeurIPS 2025 workshop
Sandbagging in a Simple Survival Bandit Problem
机构 * University of Oxford（牛津大学）
Comments Forthcoming in the "Reliable ML from Unreliable Data Workshop" at NeurIPS 2025
Ordinal Label-Distribution Learning with Constrained Asymmetric Priors for Imbalanced Retinal Grading
Comments Accepted at 39th Conference on Neural Information Processing Systems (NeurIPS 2025) Workshop: The Second Workshop on GenAI for Health: Potential, Trust, and Policy Compliance
Vocabulary Customization for Efficient Domain-Specific LLM Deployment
机构 * eBay Inc（eBay公司）
Comments Accepted at NEURIPS 2025 CCFM Workshop
Fading to Grow: Growing Preference Ratios via Preference Fading Discrete Diffusion for Recommendation
Journal ref NeurIPS 2025
GeoLink: Empowering Remote Sensing Foundation Model with OpenStreetMap Data
机构 * School of Earth and Space Sciences, Peking University（地球与空间科学学院，北京大学） ； College of Urban and Environmental Sciences, Peking University（城市与环境科学学院，北京大学） ； State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation, CAS（多模态人工智能系统国家重点实验室，中国科学院自动化研究所） ； Intelligent Maintenance and Operations Systems Lab（智能维护与运营系统实验室）
Comments NeurIPS 2025
The Impact of Scaling Training Data on Adversarial Robustness
机构 * ETH-Zurich（苏黎世联邦理工学院）
Comments Accepted at the workshop Reliable ML from Unreliable Data at NeurIPS 2025
RL-Guided Data Selection for Language Model Finetuning
机构 * Stanford University（斯坦福大学）
Comments To appear in NeurIPS 2025 Constrained Optimization for ML Workshop
MIDAS: Misalignment-based Data Augmentation Strategy for Imbalanced Multimodal Learning
机构 * KAIST（韩国科学技术院）
Comments Accepted to NeurIPS 2025
Boundary-to-Region Supervision for Offline Safe Reinforcement Learning
机构 * Harbin Institute of Technology, Weihai（哈尔滨工业大学（威海）） ； Harbin Institute of Technology, Harbin（哈尔滨工业大学） ； Westlake University, Hangzhou（西湖大学）
Comments NeurIPS 2025
Generalized Contrastive Learning for Universal Multimodal Retrieval
机构 * Qualcomm AI Research（高通人工智能研究）
Comments Accepted to NeurIPS 2025
Echoes of Humanity: Exploring the Perceived Humanness of AI Music
机构 * Universidade Federal de Minas Gerais (UFMG)（巴西联邦大学矿务格里斯大学）
Comments Accepted at NeuRIPs 2025 Creative AI Track
Ethics Statements in AI Music Papers: The Effective and the Ineffective
机构 * Northwestern University（西北大学）
Comments Accepted to the 39th Conference on Neural Information Processing Systems (NeurIPS 2025) Workshop: AI for Music
Joint Embeddings Go Temporal
机构 * KTH Stockholm（瑞典斯德哥尔摩皇家理工学院） ； New York University（纽约大学） ； Flatiron Institute（Flatiron研究所）
Comments Accepted at the Workshop on Time Series in the Age of Large Models - NeurIPS 2024
Safe and Stable Control via Lyapunov-Guided Diffusion Models
Comments Accepted by NeurIPS 2025
MAESTRO : Adaptive Sparse Attention and Robust Learning for Multimodal Dynamic Time Series
机构 * Northwestern University（西北大学）
Comments Accepted to Neurips 2025 (Spotlight)
MMPB: It's Time for Multi-Modal Personalization
机构 * AIDAS Laboratory（AIDAS实验室） ； IPAI ； ECE（电子工程系） ； Seoul National University（首尔国立大学）
Comments Accepted in NeurIPS 2025
OrthoLoC: UAV 6-DoF Localization and Calibration Using Orthographic Geodata
机构 * DeepScenario ； TU Munich（慕尼黑工业大学） ； Munich Center of Machine Learning（慕尼黑机器学习中心）
Comments Accepted at NeurIPS 2025
Scaling RL to Long Videos
机构 * NVIDIA ； MIT（麻省理工学院） ； HKU（香港大学） ； UC Berkeley（加州大学伯克利分校）
Comments Accepted by NeurIPS 2025. Code at https://github.com/NVlabs/Long-RL and model at https://huggingface.co/Efficient-Large-Model/LongVILA-R1-7B
