DICE: Discrete Interpretable Comparative Evaluation with Probabilistic Scoring for Retrieval-Augmented Generation
DICE:基于概率评分的离散可解释比较评估用于检索增强生成
机构 * School of Computer Science and Technology(计算机科学与技术学院) ; Huazhong University of Science and Technology(华中科技大学)
专题命中 检索器与排序 :retrieval-augmented generation(title,abstract);RAG(abstract);分类 cs.IR、cs.AI
AI总结 DICE提出一种基于概率评分的两阶段框架,提升RAG评估的可解释性和稳健性,通过透明判断和系统性错误诊断,实现高效且可信的评估。
Comments Accepted at ResponsibleFM @ NeurIPS 2025