PICACO: Pluralistic In-Context Value Alignment of LLMs via Total Correlation Optimization 文章

ArXiv CS.CL2026-05-28NEWSen作者: Han Jiang, Dongyao Zhu, Xiaoyuan Yi, Ziang Xiao, Zhihua Wei, Xing Xie
大语言模型
查看原文 →
关系图谱

PICACO: Pluralistic In-Context Value Alignment of LLMs via Total Correlation Optimization · 相关技术

相关技术

TTEODEPLALLMlanguage model后训练alignmentVerseUCTStanReferring expression comprehension (REC)PromptParts-of-Speech (POS) tagsOTANPUMITLMMlarge language modelsIn-context learningIMUICTHISForEffort Metric AttentionDiTARGANNTPU