用RLHF对齐大型语言模型的算法偏差：偏好崩溃和匹配正则化

{"title":"用RLHF对齐大型语言模型的算法偏差：偏好崩溃和匹配正则化","authors":"Jiancong Xiao, Ziniu Li, Xingyu Xie, Emily Getzen, Cong Fang, Qi Long, Weijie J. Su","doi":"10.1080/01621459.2025.2555067","DOIUrl":null,"url":null,"abstract":"","PeriodicalId":17227,"journal":{"name":"Journal of the American Statistical Association","volume":"9 1","pages":""},"PeriodicalIF":3.0000,"publicationDate":"2025-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"On the Algorithmic Bias of Aligning Large Language Models with RLHF: Preference Collapse and Matching Regularization\",\"authors\":\"Jiancong Xiao, Ziniu Li, Xingyu Xie, Emily Getzen, Cong Fang, Qi Long, Weijie J. Su\",\"doi\":\"10.1080/01621459.2025.2555067\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"\",\"PeriodicalId\":17227,\"journal\":{\"name\":\"Journal of the American Statistical Association\",\"volume\":\"9 1\",\"pages\":\"\"},\"PeriodicalIF\":3.0000,\"publicationDate\":\"2025-09-17\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of the American Statistical Association\",\"FirstCategoryId\":\"100\",\"ListUrlMain\":\"https://doi.org/10.1080/01621459.2025.2555067\",\"RegionNum\":1,\"RegionCategory\":\"数学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"STATISTICS & PROBABILITY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of the American Statistical Association","FirstCategoryId":"100","ListUrlMain":"https://doi.org/10.1080/01621459.2025.2555067","RegionNum":1,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"STATISTICS & PROBABILITY","Score":null,"Total":0}