TrustScore: Reference-Free Evaluation of LLM Response Trustworthiness

Danna Zheng, Danyang Liu, Mirella Lapata, Jeff Z. Pan

Keywords: LLM, Evaluation, Trustworthiness

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 closed-book question-answering tasks, where non-experts may struggle to identify inaccuracies due to the absence of contextual or ground truth information.
This paper introduces TrustScore, a framework based on the concept of Behavioral Consistency, which evaluates whether an LLM's response aligns with its intrinsic knowledge.
Additionally, TrustScore can seamlessly integrate with fact-checking methods, which assesses alignment with external knowledge sources.
The experimental results show that TrustScore achieves strong correlations with human judgments, surpassing existing reference-free metrics, and achieving results on par with reference-based metrics.

Submission Number: 66