@inproceedings{chang-etal-2025-chatbench,
title = "{C}hat{B}ench: From Static Benchmarks to Human-{AI} Evaluation",
author = "Chang, Serina and
Anderson, Ashton and
Hofman, Jake M.",
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.1262/",
doi = "10.18653/v1/2025.acl-long.1262",
pages = "26009--26038",
ISBN = "979-8-89176-251-0",
abstract = "With the rapid adoption of LLM-based chat-bots, there is a pressing need to evaluate what humans and LLMs can achieve together. However, standard benchmarks, such as MMLU, measure LLM capabilities in isolation (i.e., ``AI-alone''). Here, we design and conduct a user study to convert MMLU questions into user-AI conversations, by seeding the user with the question and having them carry out a conversation with the LLM to answer their question. We release ChatBench, a new dataset with AI-alone, user-alone, and user-AI data for 396 questions and two LLMs, including 144K answers and 7,336 user-AI conversations. We find that AI-alone accuracy fails to predict user-AI accuracy, with significant differences across multiple subjects (math, physics, and moral reasoning), and we analyze the user-AI conversations to provide insight into how they diverge from AI-alone benchmarks. Finally, we show that fine-tuning a user simulator on a subset of ChatBench improves its ability to estimate user-AI accuracies, increasing correlation on held-out questions by more than 20 points, creating possibilities for scaling interactive evaluation."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chang-etal-2025-chatbench">
<titleInfo>
<title>ChatBench: From Static Benchmarks to Human-AI Evaluation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Serina</namePart>
<namePart type="family">Chang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ashton</namePart>
<namePart type="family">Anderson</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jake</namePart>
<namePart type="given">M</namePart>
<namePart type="family">Hofman</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>With the rapid adoption of LLM-based chat-bots, there is a pressing need to evaluate what humans and LLMs can achieve together. However, standard benchmarks, such as MMLU, measure LLM capabilities in isolation (i.e., “AI-alone”). Here, we design and conduct a user study to convert MMLU questions into user-AI conversations, by seeding the user with the question and having them carry out a conversation with the LLM to answer their question. We release ChatBench, a new dataset with AI-alone, user-alone, and user-AI data for 396 questions and two LLMs, including 144K answers and 7,336 user-AI conversations. We find that AI-alone accuracy fails to predict user-AI accuracy, with significant differences across multiple subjects (math, physics, and moral reasoning), and we analyze the user-AI conversations to provide insight into how they diverge from AI-alone benchmarks. Finally, we show that fine-tuning a user simulator on a subset of ChatBench improves its ability to estimate user-AI accuracies, increasing correlation on held-out questions by more than 20 points, creating possibilities for scaling interactive evaluation.</abstract>
<identifier type="citekey">chang-etal-2025-chatbench</identifier>
<identifier type="doi">10.18653/v1/2025.acl-long.1262</identifier>
<location>
<url>https://aclanthology.org/2025.acl-long.1262/</url>
</location>
<part>
<date>2025-07</date>
<extent unit="page">
<start>26009</start>
<end>26038</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ChatBench: From Static Benchmarks to Human-AI Evaluation
%A Chang, Serina
%A Anderson, Ashton
%A Hofman, Jake M.
%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 chang-etal-2025-chatbench
%X With the rapid adoption of LLM-based chat-bots, there is a pressing need to evaluate what humans and LLMs can achieve together. However, standard benchmarks, such as MMLU, measure LLM capabilities in isolation (i.e., “AI-alone”). Here, we design and conduct a user study to convert MMLU questions into user-AI conversations, by seeding the user with the question and having them carry out a conversation with the LLM to answer their question. We release ChatBench, a new dataset with AI-alone, user-alone, and user-AI data for 396 questions and two LLMs, including 144K answers and 7,336 user-AI conversations. We find that AI-alone accuracy fails to predict user-AI accuracy, with significant differences across multiple subjects (math, physics, and moral reasoning), and we analyze the user-AI conversations to provide insight into how they diverge from AI-alone benchmarks. Finally, we show that fine-tuning a user simulator on a subset of ChatBench improves its ability to estimate user-AI accuracies, increasing correlation on held-out questions by more than 20 points, creating possibilities for scaling interactive evaluation.
%R 10.18653/v1/2025.acl-long.1262
%U https://aclanthology.org/2025.acl-long.1262/
%U https://doi.org/10.18653/v1/2025.acl-long.1262
%P 26009-26038
Markdown (Informal)
[ChatBench: From Static Benchmarks to Human-AI Evaluation](https://aclanthology.org/2025.acl-long.1262/) (Chang et al., ACL 2025)
ACL
- Serina Chang, Ashton Anderson, and Jake M. Hofman. 2025. ChatBench: From Static Benchmarks to Human-AI Evaluation. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 26009–26038, Vienna, Austria. Association for Computational Linguistics.