@inproceedings{fatahi-bayat-etal-2025-factbench,
title = "{F}act{B}ench: A Dynamic Benchmark for In-the-Wild Language Model Factuality Evaluation",
author = "Fatahi Bayat, Farima and
Zhang, Lechen and
Munir, Sheza and
Wang, Lu",
editor = "Che, Wanxiang and
Nabende, Joyce and
Shutova, Ekaterina and
Pilehvar, Mohammad Taher",
booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.acl-long.1587/",
doi = "10.18653/v1/2025.acl-long.1587",
pages = "33090--33110",
ISBN = "979-8-89176-251-0",
abstract = "The rapid adoption of language models (LMs) across diverse applications has raised concerns about their factuality, i.e., their consistency with real-world facts. We introduce VERIFY, an evidence-based evaluation pipeline that measures LMs' factuality in real-world user interactions. VERIFY considers the verifiability of LM-generated content and categorizes content units as Supported, Unsupported, or Undecidable based on Web-retrieved evidence. Importantly, factuality judgment by VERIFY more strongly correlates with human evaluations than existing methods. Using VERIFY, we identify ``hallucination prompts,'' i.e., those that frequently elicit factual errors in LM responses. These prompts form FactBench, a dataset of 1K prompts spanning 150 topics and tiered into Easy, Moderate, and Hard prompts. We benchmark widely-used openweight and proprietary LMs from six families, yielding three key findings: (i) LMs' factual precision declines from Easy to Hard prompts, (ii) factuality does not necessarily improve with scale; Llama3.1-405B-Instruct performs comparably to or worse than its 70B variant, and (iii) Gemini1.5-Pro shows a notably higher refusal rate, with over-refusal in 25{\%} of cases."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="fatahi-bayat-etal-2025-factbench">
<titleInfo>
<title>FactBench: A Dynamic Benchmark for In-the-Wild Language Model Factuality Evaluation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Farima</namePart>
<namePart type="family">Fatahi Bayat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lechen</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sheza</namePart>
<namePart type="family">Munir</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wanxiang</namePart>
<namePart type="family">Che</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joyce</namePart>
<namePart type="family">Nabende</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Shutova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohammad</namePart>
<namePart type="given">Taher</namePart>
<namePart type="family">Pilehvar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Vienna, Austria</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-251-0</identifier>
</relatedItem>
<abstract>The rapid adoption of language models (LMs) across diverse applications has raised concerns about their factuality, i.e., their consistency with real-world facts. We introduce VERIFY, an evidence-based evaluation pipeline that measures LMs’ factuality in real-world user interactions. VERIFY considers the verifiability of LM-generated content and categorizes content units as Supported, Unsupported, or Undecidable based on Web-retrieved evidence. Importantly, factuality judgment by VERIFY more strongly correlates with human evaluations than existing methods. Using VERIFY, we identify “hallucination prompts,” i.e., those that frequently elicit factual errors in LM responses. These prompts form FactBench, a dataset of 1K prompts spanning 150 topics and tiered into Easy, Moderate, and Hard prompts. We benchmark widely-used openweight and proprietary LMs from six families, yielding three key findings: (i) LMs’ factual precision declines from Easy to Hard prompts, (ii) factuality does not necessarily improve with scale; Llama3.1-405B-Instruct performs comparably to or worse than its 70B variant, and (iii) Gemini1.5-Pro shows a notably higher refusal rate, with over-refusal in 25% of cases.</abstract>
<identifier type="citekey">fatahi-bayat-etal-2025-factbench</identifier>
<identifier type="doi">10.18653/v1/2025.acl-long.1587</identifier>
<location>
<url>https://aclanthology.org/2025.acl-long.1587/</url>
</location>
<part>
<date>2025-07</date>
<extent unit="page">
<start>33090</start>
<end>33110</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T FactBench: A Dynamic Benchmark for In-the-Wild Language Model Factuality Evaluation
%A Fatahi Bayat, Farima
%A Zhang, Lechen
%A Munir, Sheza
%A Wang, Lu
%Y Che, Wanxiang
%Y Nabende, Joyce
%Y Shutova, Ekaterina
%Y Pilehvar, Mohammad Taher
%S Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria
%@ 979-8-89176-251-0
%F fatahi-bayat-etal-2025-factbench
%X The rapid adoption of language models (LMs) across diverse applications has raised concerns about their factuality, i.e., their consistency with real-world facts. We introduce VERIFY, an evidence-based evaluation pipeline that measures LMs’ factuality in real-world user interactions. VERIFY considers the verifiability of LM-generated content and categorizes content units as Supported, Unsupported, or Undecidable based on Web-retrieved evidence. Importantly, factuality judgment by VERIFY more strongly correlates with human evaluations than existing methods. Using VERIFY, we identify “hallucination prompts,” i.e., those that frequently elicit factual errors in LM responses. These prompts form FactBench, a dataset of 1K prompts spanning 150 topics and tiered into Easy, Moderate, and Hard prompts. We benchmark widely-used openweight and proprietary LMs from six families, yielding three key findings: (i) LMs’ factual precision declines from Easy to Hard prompts, (ii) factuality does not necessarily improve with scale; Llama3.1-405B-Instruct performs comparably to or worse than its 70B variant, and (iii) Gemini1.5-Pro shows a notably higher refusal rate, with over-refusal in 25% of cases.
%R 10.18653/v1/2025.acl-long.1587
%U https://aclanthology.org/2025.acl-long.1587/
%U https://doi.org/10.18653/v1/2025.acl-long.1587
%P 33090-33110
Markdown (Informal)
[FactBench: A Dynamic Benchmark for In-the-Wild Language Model Factuality Evaluation](https://aclanthology.org/2025.acl-long.1587/) (Fatahi Bayat et al., ACL 2025)
ACL