To Steer or Not to Steer? Mechanistic Error Reduction with Abstention for Language Models
机构 * J.P. Morgan AI Research(摩根大通AI研究)
Comments ICML 2025, 22 pages, 16 figures, 5 tables
Journal ref International Machine Learning Conference (ICML) 2025
期刊&会议
International Conference on Machine Learning · 会议 · Machine Learning
机构 * J.P. Morgan AI Research(摩根大通AI研究)
Comments ICML 2025, 22 pages, 16 figures, 5 tables
Journal ref International Machine Learning Conference (ICML) 2025
机构 * Department of Statistics and Data Science, Northwestern University, Evanston, USA(统计学与数据科学系,西北大学,埃文斯顿,美国) ; Department of Mechanical Engineering, Northwestern University, Evanston, USA(机械工程系,西北大学,埃文斯顿,美国) ; Department of Neurology, Northwestern University, Evanston, USA(神经病学系,西北大学,埃文斯顿,美国)
Comments 19 pages, ICML 2025
Journal ref Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:41791-41806, 2025
机构 * University of Southern California(南加州大学)
Comments ICML 2025; https://papercopilot.com/