--- title: "Dunhuang-Bench — paper by Yuan et al. (2026)" description: "The authors built Dunhuang-Bench from two expert reference works, the Dunhuang Art Dictionary and the Dunhuang Studies… 2 references." authors: ["Junyi Yuan", "Jian Zhang", "Tianxiu Yu", "Yanlin Zhou", "Xiaobo Jin", "Qiufeng Wang", "Fangyu Wu"] year: 2026 venue: "Findings of the Association for Computational Linguistics: ACL 2026" source: "https://aclanthology.org/2026.findings-acl.888/" doi: "10.18653/v1/2026.findings-acl.888" licence: "CC BY 4.0 (ACL Anthology)" licence_url: "https://creativecommons.org/licenses/by/4.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/en/papers/yuan-2026-dunhuang-bench-how-well-do/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/yuan-2026-dunhuang-bench-how-well-do/) [home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/30/) / Dunhuang-Bench: How Well Do MLLMs Understand Cultural Heritage? # Dunhuang-Bench: How Well Do MLLMs Understand Cultural Heritage? **Authors:** Junyi Yuan, [Jian Zhang](https://md.research.elamur.ai/free/en/authors/jian-zhang/), Tianxiu Yu, Yanlin Zhou, Xiaobo Jin, [Qiufeng Wang](https://md.research.elamur.ai/free/en/authors/qiufeng-wang/) and Fangyu Wu · **Year:** 2026 · **Venue:** [Findings of the Association for Computational Linguistics: ACL 2026](https://md.research.elamur.ai/free/en/venues/findings/) ([Findings of the Association for Computational Linguistics 2026](https://md.research.elamur.ai/free/en/venues/findings/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/) **Links:** [Publisher's page](https://aclanthology.org/2026.findings-acl.888/) · [PDF](https://aclanthology.org/2026.findings-acl.888.pdf) · [DOI 10.18653/v1/2026.findings-acl.888](https://doi.org/10.18653/v1/2026.findings-acl.888) · [OpenAlex](https://openalex.org/W7166844264) **Citations:** 0 citations (OpenAlex, 2026-10-14) ## Summary Multimodal large language models (MLLMs) do well on general vision–language benchmarks, but their grasp of Dunhuang art had not been tested. This art, from the Mogao Caves and other Silk Road grottoes, requires both close visual inspection and specialist cultural knowledge. The authors built Dunhuang-Bench from two expert reference works, the Dunhuang Art Dictionary and the Dunhuang Studies Dictionary. They generated questions with templates and validated them in two stages: first with GPT-5, then by humans. The benchmark is organized around art-historical theory. Panofsky's iconology underpins two tasks: visual perception (free-text QA) and knowledge reasoning (multi-turn dialogue). The formal analytic tradition underpins a third task on artistic appreciation (multiple choice). Scoring uses an LLM judge for the dialogues and a Normalized Rank Score for the appreciation task. The authors tested 20 open and closed MLLMs. Scores fall consistently from perception and appreciation down to reasoning, and even the strongest model averages only around half. Chain-of-Thought and few-shot prompting bring little gain and sometimes hurt. The resource exposes clear gaps in how MLLMs understand cultural heritage. ## Key points - Scale: 486 images and 22,970 QA pairs, about twice the size of CVQA. They split into 11,688 perception questions (Task 1), 2,020 dialogues containing 8,080 QA pairs (Task 2), and 3,202 appreciation MCQs (Task 3). - Data pipeline: 5,780 stylistic terms were taken from the Dunhuang Studies Dictionary. Text was extracted with Dolphin OCR and corrected by Dunhuang specialists. Task generation used Doubao. - Models evaluated include GPT-5.1, Gemini-2.5, and the InternVL and Qwen-VL families. The best performer, InternVL3.5-241B-A28B, averages roughly 50.9%. - Chain-of-Thought often lowers scores. For example, Qwen2.5-VL-7B drops from 12.55% to 9.85% on Task 2 and from 36.15% to 35.47% on Task 1. - Few-shot prompting mainly helps perception: Qwen2.5-VL-7B rises from 36.15% to 39.40% on Task 1 with 4 shots. It gives no reliable benefit on reasoning or appreciation. *Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.* ## Institutions - [Xi’an Jiaotong-Liverpool University](https://md.research.elamur.ai/free/en/universities/xi-an-jiaotong-liverpool-university/) — University, Suzhou, China · 11 works · 319 citations Also on the byline: [Dunhuang Research Academy](https://ror.org/05j1pnm38). 📚 Shelves (5): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 17,450 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 17,296 works · [👁️ Computer Vision and Pattern Recognition](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) · 3,486 works ## Cite Cite as: Junyi Yuan, Jian Zhang, Tianxiu Yu, Yanlin Zhou, Xiaobo Jin, Qiufeng Wang and Fangyu Wu. 2026. Dunhuang-Bench: How Well Do MLLMs Understand Cultural Heritage?. Findings of the Association for Computational Linguistics: ACL 2026. The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @inproceedings{yuan2026dunhuang, title = {{Dunhuang-Bench: How Well Do MLLMs Understand Cultural Heritage?}}, author = {Junyi Yuan, Jian Zhang, Tianxiu Yu, Yanlin Zhou, Xiaobo Jin, Qiufeng Wang and Fangyu Wu}, booktitle = {Findings of the Association for Computational Linguistics: ACL 2026}, year = {2026}, doi = {10.18653/v1/2026.findings-acl.888}, url = {https://doi.org/10.18653/v1/2026.findings-acl.888} } ``` - [← previous Duluth at SemEval-2026 Task 6: DeBERTa with LLM-Augmented Data for Unmasking Political Question Evasions](https://md.research.elamur.ai/free/en/papers/syed-2026-duluth-semeval-2026-task-6/) - [next → DVMap: Fine-Grained Pluralistic Value Alignment via High-Consensus Demographic-Value Mapping](https://md.research.elamur.ai/free/en/papers/zhu-2026-dvmap-fine-grained-pluralistic-value/) - [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/30/) ## References 2 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, and 1 others. 2023. [Gpt-4 technical report](https://md.research.elamur.ai/free/en/papers/openai-2023-gpt-4-technical-report/). arXiv preprint arXiv:2303.08774. 2. Rudolf Arnheim. 1954. Art and visual perception: A psychology of the creative eye. Univ of California Press. [find](https://scholar.google.com/scholar?q=Rudolf+Arnheim%2E+1954%2E+Art+and+visual+perception%3A+A+psychology+of+the+creative+eye%2E+Univ+of+California+Press%2E) *Updated 2 November 2026*