@inproceedings{zhou-etal-2025-essence,
title = "The Essence of Contextual Understanding in Theory of Mind: A Study on Question Answering with Story Characters",
author = "Zhou, Chulun and
Wang, Qiujing and
Yu, Mo and
Yue, Xiaoqian and
Lu, Rui and
Li, Jiangnan and
Zhou, Yifan and
Zhang, Shunchi and
Zhou, Jie and
Lam, Wai",
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.1103/",
doi = "10.18653/v1/2025.acl-long.1103",
pages = "22612--22631",
ISBN = "979-8-89176-251-0",
abstract = "Theory-of-Mind (ToM) is a fundamental psychological capability that allows humans to understand and interpret the mental states of others. Humans infer others' thoughts by integrating causal cues and indirect clues from broad contextual information, often derived from past interactions. In other words, human ToM heavily relies on the understanding about the backgrounds and life stories of others. Unfortunately, this aspect is largely overlooked in existing benchmarks for evaluating machines' ToM capabilities, due to their usage of short narratives without global context, especially personal background of characters. In this paper, we verify the importance of comprehensive contextual understanding about personal backgrounds in ToM and assess the performance of LLMs in such complex scenarios. To achieve this, we introduce CharToM-QA benchmark, comprising 1,035 ToM questions based on characters from classic novels. Our human study reveals a significant disparity in performance: the same group of educated participants performs dramatically better when they have read the novels compared to when they have not. In parallel, our experiments on state-of-the-art LLMs, including the very recent o1 and DeepSeek-R1 models, show that LLMs still perform notably worse than humans, despite that they have seen these stories during pre-training. This highlights the limitations of current LLMs in capturing the nuanced contextual information required for ToM reasoning."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhou-etal-2025-essence">
<titleInfo>
<title>The Essence of Contextual Understanding in Theory of Mind: A Study on Question Answering with Story Characters</title>
</titleInfo>
<name type="personal">
<namePart type="given">Chulun</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qiujing</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mo</namePart>
<namePart type="family">Yu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaoqian</namePart>
<namePart type="family">Yue</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rui</namePart>
<namePart type="family">Lu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiangnan</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yifan</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shunchi</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jie</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wai</namePart>
<namePart type="family">Lam</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>Theory-of-Mind (ToM) is a fundamental psychological capability that allows humans to understand and interpret the mental states of others. Humans infer others’ thoughts by integrating causal cues and indirect clues from broad contextual information, often derived from past interactions. In other words, human ToM heavily relies on the understanding about the backgrounds and life stories of others. Unfortunately, this aspect is largely overlooked in existing benchmarks for evaluating machines’ ToM capabilities, due to their usage of short narratives without global context, especially personal background of characters. In this paper, we verify the importance of comprehensive contextual understanding about personal backgrounds in ToM and assess the performance of LLMs in such complex scenarios. To achieve this, we introduce CharToM-QA benchmark, comprising 1,035 ToM questions based on characters from classic novels. Our human study reveals a significant disparity in performance: the same group of educated participants performs dramatically better when they have read the novels compared to when they have not. In parallel, our experiments on state-of-the-art LLMs, including the very recent o1 and DeepSeek-R1 models, show that LLMs still perform notably worse than humans, despite that they have seen these stories during pre-training. This highlights the limitations of current LLMs in capturing the nuanced contextual information required for ToM reasoning.</abstract>
<identifier type="citekey">zhou-etal-2025-essence</identifier>
<identifier type="doi">10.18653/v1/2025.acl-long.1103</identifier>
<location>
<url>https://aclanthology.org/2025.acl-long.1103/</url>
</location>
<part>
<date>2025-07</date>
<extent unit="page">
<start>22612</start>
<end>22631</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T The Essence of Contextual Understanding in Theory of Mind: A Study on Question Answering with Story Characters
%A Zhou, Chulun
%A Wang, Qiujing
%A Yu, Mo
%A Yue, Xiaoqian
%A Lu, Rui
%A Li, Jiangnan
%A Zhou, Yifan
%A Zhang, Shunchi
%A Zhou, Jie
%A Lam, Wai
%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 zhou-etal-2025-essence
%X Theory-of-Mind (ToM) is a fundamental psychological capability that allows humans to understand and interpret the mental states of others. Humans infer others’ thoughts by integrating causal cues and indirect clues from broad contextual information, often derived from past interactions. In other words, human ToM heavily relies on the understanding about the backgrounds and life stories of others. Unfortunately, this aspect is largely overlooked in existing benchmarks for evaluating machines’ ToM capabilities, due to their usage of short narratives without global context, especially personal background of characters. In this paper, we verify the importance of comprehensive contextual understanding about personal backgrounds in ToM and assess the performance of LLMs in such complex scenarios. To achieve this, we introduce CharToM-QA benchmark, comprising 1,035 ToM questions based on characters from classic novels. Our human study reveals a significant disparity in performance: the same group of educated participants performs dramatically better when they have read the novels compared to when they have not. In parallel, our experiments on state-of-the-art LLMs, including the very recent o1 and DeepSeek-R1 models, show that LLMs still perform notably worse than humans, despite that they have seen these stories during pre-training. This highlights the limitations of current LLMs in capturing the nuanced contextual information required for ToM reasoning.
%R 10.18653/v1/2025.acl-long.1103
%U https://aclanthology.org/2025.acl-long.1103/
%U https://doi.org/10.18653/v1/2025.acl-long.1103
%P 22612-22631
Markdown (Informal)
[The Essence of Contextual Understanding in Theory of Mind: A Study on Question Answering with Story Characters](https://aclanthology.org/2025.acl-long.1103/) (Zhou et al., ACL 2025)
ACL
- Chulun Zhou, Qiujing Wang, Mo Yu, Xiaoqian Yue, Rui Lu, Jiangnan Li, Yifan Zhou, Shunchi Zhang, Jie Zhou, and Wai Lam. 2025. The Essence of Contextual Understanding in Theory of Mind: A Study on Question Answering with Story Characters. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 22612–22631, Vienna, Austria. Association for Computational Linguistics.