@inproceedings{xu-etal-2026-tcc,
title = "{TCC}-Bench: Benchmarking the Traditional {C}hinese Culture Understanding Capabilities of {MLLM}s",
author = "Xu, Pengju and
Wang, Yan and
Zhang, Shuyuan and
Zhou, Xuan and
Lixin and
Yuanyue and
Li, Fengzhao and
Zhou, Shunyuan and
Wang, Xingyu and
Zhang, Yi and
Zhao, Haiying",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.1219/",
pages = "27834--27854",
ISBN = "979-8-89176-459-0",
abstract = "Recent progress in Multimodal Large Language Models (MLLMs) has significantly enhanced the ability of artificial intelligence systems to understand and generate multimodal content. However, the capabilities of existing MLLMs in understanding traditional Chinese culture have not yet been comprehensively explored. To address this gap, we propose the \textbf{T}raditional \textbf{C}hinese \textbf{C}ulture understanding \textbf{Bench}mark (\textbf{TCC-Bench}), a bilingual Visual Question Answering (VQA) benchmark designed to assess MLLMs' performance on culturally grounded visual content. TCC-Bench features culturally rich and visually diverse data, covering museum artifacts, daily life scenes, comics, and other culturally significant contexts across eight cultural domains. Through extensive experiments on a wide range of MLLMs, we find that current models still face substantial challenges in handling traditional Chinese cultural content. Our results highlight the need for more culturally inclusive and context-aware multimodal systems. The code and data can be found at: \url{https://github.com/Morty-Xu/TCC-Bench}."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="xu-etal-2026-tcc">
<titleInfo>
<title>TCC-Bench: Benchmarking the Traditional Chinese Culture Understanding Capabilities of MLLMs</title>
</titleInfo>
<name type="personal">
<namePart type="given">Pengju</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yan</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shuyuan</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xuan</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name>
<namePart>Lixin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name>
<namePart>Yuanyue</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fengzhao</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shunyuan</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xingyu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yi</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haiying</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Recent progress in Multimodal Large Language Models (MLLMs) has significantly enhanced the ability of artificial intelligence systems to understand and generate multimodal content. However, the capabilities of existing MLLMs in understanding traditional Chinese culture have not yet been comprehensively explored. To address this gap, we propose the Traditional Chinese Culture understanding Benchmark (TCC-Bench), a bilingual Visual Question Answering (VQA) benchmark designed to assess MLLMs’ performance on culturally grounded visual content. TCC-Bench features culturally rich and visually diverse data, covering museum artifacts, daily life scenes, comics, and other culturally significant contexts across eight cultural domains. Through extensive experiments on a wide range of MLLMs, we find that current models still face substantial challenges in handling traditional Chinese cultural content. Our results highlight the need for more culturally inclusive and context-aware multimodal systems. The code and data can be found at: https://github.com/Morty-Xu/TCC-Bench.</abstract>
<identifier type="citekey">xu-etal-2026-tcc</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.1219/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>27834</start>
<end>27854</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T TCC-Bench: Benchmarking the Traditional Chinese Culture Understanding Capabilities of MLLMs
%A Xu, Pengju
%A Wang, Yan
%A Zhang, Shuyuan
%A Zhou, Xuan
%A Li, Fengzhao
%A Zhou, Shunyuan
%A Wang, Xingyu
%A Zhang, Yi
%A Zhao, Haiying
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%A Lixin
%A Yuanyue
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F xu-etal-2026-tcc
%X Recent progress in Multimodal Large Language Models (MLLMs) has significantly enhanced the ability of artificial intelligence systems to understand and generate multimodal content. However, the capabilities of existing MLLMs in understanding traditional Chinese culture have not yet been comprehensively explored. To address this gap, we propose the Traditional Chinese Culture understanding Benchmark (TCC-Bench), a bilingual Visual Question Answering (VQA) benchmark designed to assess MLLMs’ performance on culturally grounded visual content. TCC-Bench features culturally rich and visually diverse data, covering museum artifacts, daily life scenes, comics, and other culturally significant contexts across eight cultural domains. Through extensive experiments on a wide range of MLLMs, we find that current models still face substantial challenges in handling traditional Chinese cultural content. Our results highlight the need for more culturally inclusive and context-aware multimodal systems. The code and data can be found at: https://github.com/Morty-Xu/TCC-Bench.
%U https://aclanthology.org/2026.emnlp-main.1219/
%P 27834-27854
Markdown (Informal)
[TCC-Bench: Benchmarking the Traditional Chinese Culture Understanding Capabilities of MLLMs](https://aclanthology.org/2026.emnlp-main.1219/) (Xu et al., EMNLP 2026)
ACL
- Pengju Xu, Yan Wang, Shuyuan Zhang, Xuan Zhou, Lixin, Yuanyue, Fengzhao Li, Shunyuan Zhou, Xingyu Wang, Yi Zhang, and Haiying Zhao. 2026. TCC-Bench: Benchmarking the Traditional Chinese Culture Understanding Capabilities of MLLMs. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 27834–27854, Budapest, Hungary. Association for Computational Linguistics.