Unlocking the Pre-Trained Model as a Dual-Alignment Calibrator for Post-Trained LLMs
解封预训练模型作为后训练LLMs的双对齐校准器
机构 * Department of Statistics and Data Science, Southern University of Science and Technology(统计与数据科学系,南方科技大学) ; School of Computing, National University of Singapore(计算学院,新加坡国立大学) ; Department of Computer Sciences, University of Wisconsin-Madison(计算机科学系,威斯康星大学麦迪逊分校) ; College of Computing and Data Science, Nanyang Technological University(计算与数据科学学院,南洋理工大学)
专题命中 幻觉与事实性 :alignment(title,abstract);分类 cs.LG
AI总结 本文提出Dual-Align方法,通过双对齐策略校正后训练LLMs的置信度漂移和过程漂移,提升校准性能。