HaluEval: A Large-Scale Hallucination Evaluation Benchmark for Large Language Models

By Junyi Li, Xiaoxue Cheng, Wayne Xin Zhao, Jian-Yun Nie, Ji-rong Wen. Conference on Empirical Methods in Natural Language Processing, 2023

504 citations recorded by the source corpus.

Large language models (LLMs), such as ChatGPT, are prone to generate hallucinations, i.e., content that conflicts with the source or cannot be verified by the factual knowledge. To understand what types of content and to which extent LLMs are apt to hallucinate, we introduce the Hallucination Evaluation benchmark for Large Language Models (HaluEval), a large collection of generated and human-annotated hallucinated samples for evaluating the performance of LLMs in recognizing hallucination. To generate these samples, we propose a ChatGPT-based two-step framework, i.e., sampling-then-filtering. Besides, we also hire some human labelers to annotate the hallucinations in ChatGPT responses. The empirical results suggest that ChatGPT is likely to generate hallucinated content in specific topics by fabricating unverifiable information (i.e., about $19.5\%$ responses). Moreover, existing LLMs face great challenges in recognizing the hallucinations in texts. However, our experiments also prove that providing external knowledge or adding reasoning steps can help LLMs recognize hallucinations. Our benchmark can be accessed at https://github.com/RUCAIBox/HaluEval.

Subfields: Artificial Intelligence, Social Psychology

View Paper: https://www.semanticscholar.org/paper/e0384ba36555232c587d4a80d527895a095a9001

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning (2024, background): “In response, safety benchmarks are used to monitor and mitigate these behaviors (Hendrycks et al., 2020a; Lin et al., 2021; Li et al., 2023; Pan et al., 2023; Kinniment and Sato, 2023; Inan et al., 2023).”
Large Language Models: A Survey (2024, unknown): “HaluEval is one of the datasets that aims to measure hallucination in this field [205].”
Hallucination is Inevitable: An Innate Limitation of Large Language Models (2024, background): “For example, factual-centred metrics [19, 20, 40, 57] and benchmarks [34, 35, 65] have been proposed to measure and reduce hallucination on specific datasets.”
Large Language Model Unlearning (2023, methodology): “We select the hallucinated Q&A pairs (i.e. negative samples) in the HaluE-val [Li et al., 2023] dataset as D fgt and TruthfulQA [Lin et al., 2021] dataset as D nor .”
Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Models (2023, unknown): “, 2022), or factual knowledge (Min et al., 2023; Muhlgay et al., 2023; Li et al., 2023a)—this phenomenon is commonly referred to as hallucination, which significantly undermines the reliability of LLMs in real-world scenarios.”