FaithBench: A Diverse Hallucination Benchmark for Summarization by Modern LLMs
Forrest Sheng Bao, Miaoran Li, Renyi Qu, Ge Luo, Erana Wan, Yujia Tang, Weisi Fan, Manveer Singh Tamber
Abstract
Summarization is one of the most common tasks performed by large language models (LLMs), especially in applications like Retrieval-Augmented Generation (RAG). However, existing evaluations of hallucinations in LLM-generated summaries, and evaluations of hallucination detection models both suffer from a lack of diversity and recency in the LLM and LLM families considered. This paper introduces FaithBench, a summarization hallucination benchmark comprising challenging hallucinations made by 10 modern LLMs from 8 different families, with ground truth annotations by human experts. “Challenging” here means summaries on which popular, state-of-the-art hallucination detection models, including GPT-4o-as-a-judge, disagreed on. Our results show GPT-4o and GPT-3.5-Turbo produce the least hallucinations. However, most state-of-the-art hallucination detection models have near 50% accuracies on FaithBench, indicating lots of room for future improvement.
BibTeX
@inproceedings{bao-etal-2025-faithbench,
title = "{F}aith{B}ench: A Diverse Hallucination Benchmark for Summarization by {M}odern {LLM}s",
author = "Bao, Forrest Sheng and
Li, Miaoran and
Qu, Renyi and
Luo, Ge and
Wan, Erana and
Tang, Yujia and
Fan, Weisi and
Tamber, Manveer Singh and
Kazi, Suleman and
Sourabh, Vivek and
Qi, Mike and
Tu, Ruixuan and
Xu, Chenyu and
Gonzales, Matthew and
Mendelevitch, Ofer and
Ahmad, Amin",
editor = "Chiruzzo, Luis and
Ritter, Alan and
Wang, Lu",
booktitle = "Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 2: Short Papers)",
month = apr,
year = "2025",
address = "Albuquerque, New Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.naacl-short.38/",
pages = "448--461",
ISBN = "979-8-89176-190-2"
}