A New Benchmark and Reverse Validation Method for Passage-level Hallucination Detection

TL;DR

This paper introduces a new benchmark called PHD and a reverse validation method for passage-level hallucination detection in LLMs, demonstrating improved accuracy and efficiency over existing zero-resource approaches.

Contribution

It presents a novel passage-level hallucination detection benchmark and a self-check method based on reverse validation, advancing zero-resource detection techniques.

Findings

The proposed method outperforms baselines in accuracy.

It requires fewer tokens and less time.

Shared limitations of zero-resource methods are identified.

Abstract

Large Language Models (LLMs) have shown their ability to collaborate effectively with humans in real-world scenarios. However, LLMs are apt to generate hallucinations, i.e., makeup incorrect text and unverified information, which can cause significant damage when deployed for mission-critical tasks. In this paper, we propose a self-check approach based on reverse validation to detect factual errors automatically in a zero-resource fashion. To facilitate future studies and assess different methods, we construct a hallucination detection benchmark named PHD, which is generated by ChatGPT and annotated by human annotators. Contrasting previous studies of zero-resource hallucination detection, our method and benchmark concentrate on passage-level detection instead of sentence-level. We empirically evaluate our method and existing zero-resource detection methods on two datasets. The experimental results demonstrate that the proposed method considerably outperforms the baselines while costing fewer tokens and less time. Furthermore, we manually analyze some hallucination cases that LLM failed to capture, revealing the shared limitation of zero-resource methods.