BriefGPT.xyz
Feb, 2024
TrustScore: 无需参考的LLM响应可信度评估
TrustScore: Reference-Free Evaluation of LLM Response Trustworthiness
HTML
PDF
Danna Zheng, Danyang Liu, Mirella Lapata, Jeff Z. Pan
TL;DR
本研究提出了基于行为一致性概念的TrustScore框架，用于评估大型语言模型（LLMs）的响应与其内在知识的一致性，同时能够与事实核实方法无缝集成，实现与人类判断强相关性的结果。
Abstract
large language models
(LLMs) have demonstrated impressive capabilities across various domains, prompting a surge in their practical applications. However, concerns have arisen regarding the trustworthiness of LLMs outputs, particularly in
→