@inproceedings{wang-etal-2025-ubench,
title = "{UB}ench: Benchmarking Uncertainty in Large Language Models with Multiple Choice Questions",
author = "Wang, Xunzhi and
Zhang, Zhuowei and
Chen, Gaonan and
Li, Qiongyu and
Luo, Bitong and
Han, Zhixin and
Wang, Haotian and
Li, Zhiyu and
Gao, Hang and
Hu, Mengting",
editor = "Che, Wanxiang and
Nabende, Joyce and
Shutova, Ekaterina and
Pilehvar, Mohammad Taher",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2025",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-acl.423/",
doi = "10.18653/v1/2025.findings-acl.423",
pages = "8076--8107",
ISBN = "979-8-89176-256-5",
abstract = "Despite recent progress in systematic evaluation frameworks, benchmarking the uncertainty of large language models (LLMs) remains a highly challenging task. Existing methods for benchmarking the uncertainty of LLMs face three key challenges: the need for internal model access, additional training, or high computational costs. This is particularly unfavorable for closed-source models. To this end, we introduce UBench, a new benchmark for evaluating the uncertainty of LLMs. Unlike other benchmarks, UBench is based on confidence intervals. It encompasses 11,978 multiple-choice questions spanning knowledge, language, understanding, and reasoning capabilities. Based on this, we conduct extensive experiments. This includes comparisons with other advanced uncertainty estimation methods, the assessment of the uncertainty of 20 LLMs, and an exploration of the effects of Chain-of-Thought (CoT) prompts, role-playing (RP) prompts, and temperature on model uncertainty. Our analysis reveals several crucial insights: 1) Our confidence interval-based methods are highly effective for uncertainty quantification; 2) Regarding uncertainty, outstanding open-source models show competitive performance versus closed-source models; 3) CoT and RP prompts present potential ways to improve model reliability, while the influence of temperature changes follows no universal rule. Our implementation is available at https://github.com/Cyno2232/UBENCH."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2025-ubench">
<titleInfo>
<title>UBench: Benchmarking Uncertainty in Large Language Models with Multiple Choice Questions</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xunzhi</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhuowei</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gaonan</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qiongyu</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bitong</namePart>
<namePart type="family">Luo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhixin</namePart>
<namePart type="family">Han</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haotian</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhiyu</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hang</namePart>
<namePart type="family">Gao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mengting</namePart>
<namePart type="family">Hu</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>Findings of the Association for Computational Linguistics: ACL 2025</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">979-8-89176-256-5</identifier>
</relatedItem>
<abstract>Despite recent progress in systematic evaluation frameworks, benchmarking the uncertainty of large language models (LLMs) remains a highly challenging task. Existing methods for benchmarking the uncertainty of LLMs face three key challenges: the need for internal model access, additional training, or high computational costs. This is particularly unfavorable for closed-source models. To this end, we introduce UBench, a new benchmark for evaluating the uncertainty of LLMs. Unlike other benchmarks, UBench is based on confidence intervals. It encompasses 11,978 multiple-choice questions spanning knowledge, language, understanding, and reasoning capabilities. Based on this, we conduct extensive experiments. This includes comparisons with other advanced uncertainty estimation methods, the assessment of the uncertainty of 20 LLMs, and an exploration of the effects of Chain-of-Thought (CoT) prompts, role-playing (RP) prompts, and temperature on model uncertainty. Our analysis reveals several crucial insights: 1) Our confidence interval-based methods are highly effective for uncertainty quantification; 2) Regarding uncertainty, outstanding open-source models show competitive performance versus closed-source models; 3) CoT and RP prompts present potential ways to improve model reliability, while the influence of temperature changes follows no universal rule. Our implementation is available at https://github.com/Cyno2232/UBENCH.</abstract>
<identifier type="citekey">wang-etal-2025-ubench</identifier>
<identifier type="doi">10.18653/v1/2025.findings-acl.423</identifier>
<location>
<url>https://aclanthology.org/2025.findings-acl.423/</url>
</location>
<part>
<date>2025-07</date>
<extent unit="page">
<start>8076</start>
<end>8107</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T UBench: Benchmarking Uncertainty in Large Language Models with Multiple Choice Questions
%A Wang, Xunzhi
%A Zhang, Zhuowei
%A Chen, Gaonan
%A Li, Qiongyu
%A Luo, Bitong
%A Han, Zhixin
%A Wang, Haotian
%A Li, Zhiyu
%A Gao, Hang
%A Hu, Mengting
%Y Che, Wanxiang
%Y Nabende, Joyce
%Y Shutova, Ekaterina
%Y Pilehvar, Mohammad Taher
%S Findings of the Association for Computational Linguistics: ACL 2025
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria
%@ 979-8-89176-256-5
%F wang-etal-2025-ubench
%X Despite recent progress in systematic evaluation frameworks, benchmarking the uncertainty of large language models (LLMs) remains a highly challenging task. Existing methods for benchmarking the uncertainty of LLMs face three key challenges: the need for internal model access, additional training, or high computational costs. This is particularly unfavorable for closed-source models. To this end, we introduce UBench, a new benchmark for evaluating the uncertainty of LLMs. Unlike other benchmarks, UBench is based on confidence intervals. It encompasses 11,978 multiple-choice questions spanning knowledge, language, understanding, and reasoning capabilities. Based on this, we conduct extensive experiments. This includes comparisons with other advanced uncertainty estimation methods, the assessment of the uncertainty of 20 LLMs, and an exploration of the effects of Chain-of-Thought (CoT) prompts, role-playing (RP) prompts, and temperature on model uncertainty. Our analysis reveals several crucial insights: 1) Our confidence interval-based methods are highly effective for uncertainty quantification; 2) Regarding uncertainty, outstanding open-source models show competitive performance versus closed-source models; 3) CoT and RP prompts present potential ways to improve model reliability, while the influence of temperature changes follows no universal rule. Our implementation is available at https://github.com/Cyno2232/UBENCH.
%R 10.18653/v1/2025.findings-acl.423
%U https://aclanthology.org/2025.findings-acl.423/
%U https://doi.org/10.18653/v1/2025.findings-acl.423
%P 8076-8107
Markdown (Informal)
[UBench: Benchmarking Uncertainty in Large Language Models with Multiple Choice Questions](https://aclanthology.org/2025.findings-acl.423/) (Wang et al., Findings 2025)
ACL
- Xunzhi Wang, Zhuowei Zhang, Gaonan Chen, Qiongyu Li, Bitong Luo, Zhixin Han, Haotian Wang, Zhiyu Li, Hang Gao, and Mengting Hu. 2025. UBench: Benchmarking Uncertainty in Large Language Models with Multiple Choice Questions. In Findings of the Association for Computational Linguistics: ACL 2025, pages 8076–8107, Vienna, Austria. Association for Computational Linguistics.