Evaluating Evaluation Metrics – The Mirage of Hallucination Detection

Atharva Kulkarni, Yuan Zhang, Joel Ruben Antony Moniz, Xiou Ge, Bo-Hsiang Tseng, Dhivya Piraviperumal, Swabha Swayamdipta, Hong Yu

Keywords: Hallucination Detection, Meta Evaluation of Metrics, Dialog Systems, Resources and Evaluation, NLP Applications

TL;DR: Meta-evaluation of hallucination detection metrics for Large Language Model (LLM) generated text

Abstract: Hallucinations pose a significant obstacle to the reliability and widespread adoption of language models, yet their accurate measurement remains a persistent challenge. While many task- and domain-specific metrics have been proposed to assess faithfulness and factuality concerns, the robustness and generalization of these metrics are still untested. In this paper, we conduct a large-scale empirical evaluation of 6 diverse sets of hallucination detection metrics across 4 datasets, 37 language models from 5 families, and 5 decoding methods. Our extensive investigation reveals concerning gaps in current hallucination evaluation: metrics often fail to align with human judgments, take an overtly myopic view of the problem, and show inconsistent gains with parameter scaling. Encouragingly, LLM-based evaluation, particularly with GPT-4, yields the best overall results, and mode-seeking decoding methods seem to reduce hallucinations, especially in knowledge-grounded settings. These findings underscore the need for more robust metrics to understand and quantify hallucinations, and better strategies to mitigate them.

Submission Number: 173