LURE: Live-Usage Replay Evaluations for Reducing Evaluation Awareness 文章

ArXiv CS.CL2026-05-27NEWSen作者: Igor Ivanov, David Demitri Africa
大语言模型
查看原文 →
关系图谱

LURE: Live-Usage Replay Evaluations for Reducing Evaluation Awareness · 相关技术

相关技术

LLMODEPLAlanguage modelsafetyrsaevaldivide-and-conquer partitioningalignmentUCTTeraStanSAGESAFEReferring expression comprehension (REC)PromptParts-of-Speech (POS) tagsPPRlarge language modelsForDiTARGANNAgent