arXiv cs.AI· Maria Victoria Carro, Denise Alejandra Mester, Facundo Nieto, Oscar Agustin Stanchi, Guido Ernesto Bergman, Giovanni Franco Gabriel Marrafini, Trinidad Borrell, Guido Freire, Joaquin Salvador Machulsky, Mario Alejandro Leiva, Ana Lopez, Nicolas Spinelli, Luca Nicolas Forziati Gangi, Eitan Sprejer, Federico Barrera Lemarchand, Gerardo Simari, Maria Vanina Martinez·· 1 天前AI 评分50
辩论能帮人类纠正判断，却帮不了较弱的 AI 模型
Humans Correct Their Judgments Through Debate, but Weaker AI Models Do Not
AI 导读
一项 arXiv 研究检验「AI Safety via Debate」机制下人类与 LLM 监督者的表现：辩论使人类裁判的判断准确率相比先前信念有所提升，而单一专家陈述的 consultancy 基线则无此效果。
来源：arXiv cs.AI · arxiv.org