Escaping the SpuriVerse: Can Large Vision-Language Models Generalize Beyond Seen Spurious Correlations?
机构 * University of Washington(华盛顿大学) ; Stanford University(斯坦福大学) ; University of Michigan(密歇根大学)
高校专区
机构 * University of Washington(华盛顿大学) ; Stanford University(斯坦福大学) ; University of Michigan(密歇根大学)
机构 * University of Maryland Baltimore County(马里兰大学巴尔的摩分校) ; Stanford University and Amazon AI(斯坦福大学和亚马逊人工智能)
Comments 19 Pages, 7 Figures, 4 Tables (Note: Under Review)
机构 * Stanford University(斯坦福大学) ; Mila, Quebec AI Institute(魁北克AI研究院)
Comments Accepted at Actionable Interpretability Workshop at ICML 2025
机构 * Artificial and Mechanical Intelligence Laboratory(人工智能与机械实验室) ; Istituto Italiano di Tecnologia(意大利技术研究院) ; Dipartimento di Ingegneria Industriale(工业工程系) ; Università di Napoli Federico II(那不勒斯费德里科二世大学) ; School of Computer Science(计算机科学学院) ; University of Manchester(曼彻斯特大学) ; Dipartimento di Scienze e Tecnologie Aerospaziali(航空航天科学与技术系) ; Politecnico di Milano(米兰理工学院) ; Department of Mechanical Engineering(机械工程系) ; Stanford University(斯坦福大学)
Journal ref Communications Engineering 4, 111 (2025)
机构 * Stanford University(斯坦福大学)
Comments ICLR 2025 Workshop on Frontiers of Probabilistic Inference
机构 * Stanford University(斯坦福大学)
机构 * Google(谷歌) ; Stanford University(斯坦福大学) ; University of California, Berkeley(加州大学伯克利分校)
Journal ref Workshop on Deepfake Detection, Localization and Interpretability @ IJCAI 2025
机构 * MIT(麻省理工学院) ; Harvard University(哈佛大学) ; University of Cambridge(剑桥大学) ; Brown University(布朗大学) ; Yale University(耶鲁大学) ; Stanford University(斯坦福大学)
Comments 5 figures, 19 pages
机构 * Stanford University(斯坦福大学) ; Mila, McGill University(麦吉尔大学米尔)
机构 * Stanford University(斯坦福大学)
机构 * UC Santa Cruz(加州大学圣克ruz分校) ; Stanford University(斯坦福大学) ; UC Santa Barbara(加州大学圣芭芭拉分校)
机构 * University of Santiago de Chile(智利圣地亚哥大学) ; University of Southern California(南加州大学) ; Olin College of Engineering(奥林工程学院) ; Stanford University(斯坦福大学)
Comments 8 pages, 7 figures
Journal ref IEEE Robotics and Automation Letters ( Volume: 10, Issue: 7, July 2025)
机构 * Institute for Computational and Mathematical Engineering (ICME), Stanford University(计算与数学工程研究所(ICME),斯坦福大学) ; Amazon(亚马逊) ; Department of Mathematics, Stanford University(数学系,斯坦福大学)
Comments Published at UAI 2025
机构 * Stanford University(斯坦福大学)
机构 * Department of Biomedical Data Science, Stanford University(生物医学数据科学系,斯坦福大学) ; Department of Econometrics & Business Statistics, Monash University(计量经济学与商业统计学系,墨尔本大学) ; Departments of Biomedical Data Science and of Statistics, Stanford University(生物医学数据科学系和统计学系,斯坦福大学)
机构 * Stanford University(斯坦福大学)
机构 * Stanford University(斯坦福大学)
Comments 27 pages, 20 figures
机构 * Department of Brain and Cognitive Science, MIT(麻省理工学院脑科学与认知科学系) ; Department of Linguistics, Stanford University(斯坦福大学语言学系)
机构 * Google(谷歌) ; University of Tübingen(图宾根大学) ; Vector Institute(向量研究所) ; Dalhousie University(达尔豪斯大学) ; University of Toronto(多伦多大学) ; Stanford University(斯坦福大学) ; Meta AI (FAIR)(Meta AI(FAIR))
Comments 102 pages, 8 figures, 41 tables
机构 * University of Southern California(南加州大学) ; Stanford University(斯坦福大学)
Comments Findings of ACL 2025
机构 * Stanford University(斯坦福大学) ; NVIDIA Research(NVIDIA研究)
机构 * Carnegie Mellon University(卡内基梅隆大学) ; Fudan University(复旦大学) ; Imperial College London(伦敦帝国学院) ; Massachusetts Institute of Technology(麻省理工学院) ; National University of Singapore(新加坡国立大学) ; Peking University(北京大学) ; Shanghai Jiao Tong University(上海交通大学) ; Stanford University(斯坦福大学) ; The Chinese University of Hong Kong (Shenzhen)(香港中文大学(深圳)) ; The Ohio State University(俄亥俄州立大学) ; Tsinghua University(清华大学) ; University of Chinese Academy of Sciences(中国科学院大学) ; University of Oxford(牛津大学) ; University of Pennsylvania(宾夕法尼亚大学) ; University of Science and Technology of China(中国科学技术大学) ; University of Sydney(悉尼大学) ; University of Toronto(多伦多大学) ; Yale University(耶鲁大学)
Comments Project page: https://xbench.org
机构 * Department of Computer Science, University of California, Los Angeles(加州大学洛杉矶分校计算机科学系) ; Department of Computer Science, Stanford University(斯坦福大学计算机科学系)
机构 * Stanford University(斯坦福大学)
机构 * AI Thrust, Information Hub, HKUST(GZ)(香港科技大学(广州)人工智能 thrust 与信息中心) ; College of Computer Science and Technology, Zhejiang University(浙江大学计算机科学与技术学院) ; School of Computer Science and Engineering, South China University of Technology(华南理工大学计算机科学与工程学院) ; School of Medicine, Stanford University(斯坦福大学医学院) ; Computer Science Department, UIUC(伊利诺伊大学厄巴纳-香槟分校计算机科学系) ; Medical Big Data Center, Guangdong Provincial People’s Hospital (Guangdong Academy of Medical Sciences), Southern Medical University(广东省人民医院(广东省医学科学院)医学大数据中心,南方医科大学) ; Innovation Institute for Artificial Intelligence in Medicine of Zhejiang University, College of Pharmaceutical Sciences, Zhejiang University(浙江大学人工智能医学创新研究院,浙江大学药学院) ; The Second Affiliated Hospital, Zhejiang University School of Medicine(浙江大学医学院附属第二医院) ; GE HealthCare(通用电气医疗) ; Shanghai Artificial Intelligence Laboratory(上海人工智能实验室) ; IQVIA ; Computer Science Department, Stanford University(斯坦福大学计算机科学系) ; Informatics, Harvard Medical School, Harvard University(哈佛医学院,哈佛大学informatics部门) ; State Key Laboratory for Novel Software Technology at Nanjing University, School of Computer Science, Nanjing University(南京大学新型软件技术国家重点实验室,南京大学计算机科学学院)
Comments accepted by Nature Scientific Data
机构 * MIT(麻省理工学院) ; Stanford University(斯坦福大学) ; University College London(伦敦大学学院)
Comments For Github code see https://github.com/vdlad/Remarkable-Robustness-of-LLMs. Send all correspondence to the first author
机构 * Stanford University(斯坦福大学)
Comments 60 pages, 8 figures, accepted by Neurips 2024
机构 * University of Toronto(多伦多大学) ; Stanford University(斯坦福大学)
机构 * Stanford University(斯坦福大学) ; Massachusetts Institute of Technology(麻省理工学院)
机构 * University of California, Berkeley(加州大学伯克利分校) ; Massachusetts Institute of Technology(麻省理工学院) ; Stanford University(斯坦福大学) ; Google Research(谷歌研究)
Comments Code available at https://github.com/aangelopoulos/conformal-risk