ChinaHeritaQA: A Culturally-Grounded Visual Question Answering Dataset for World Heritage Sites in China
On the map: LLM reasoning → Multimodal reasoning
Abstract (arXiv)
We introduce ChinaHeritaQA, a multimodal benchmark dataset for evaluating the cultural reasoning abilities of vision-language models (VLMs) on UNESCO World Heritage sites in China. The dataset comprises 2,279 in-the-wild images paired with 14,133 bilingual (Chinese/English) multiple-choice QA pairs spanning seven cognitive dimensions, from basic identity recognition to historical periodization and architectural analysis. Guided by a UNESCO-aligned heritage ontology and verified through rigorous human annotation, the dataset ensures linguistic quality and factual consistency. Evaluations of state-of-the-art VLMs reveal that while top models exceed human performance on average, substantial task-level variation emerges: models excel at visual recognition but struggle with culturally grounded reasoning. Performance also varies by dynasty and region. ChinaHeritaQA reveals that strong visual retrieval does not extend to cultural and historical understanding. We release the dataset to support future research on culturally aware multimodal learning.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Multimodal reasoning, one of 17 directions in LLM reasoning. Builds on 10 of its 36 references that are in the map. First posted on arXiv 8 Jun 2026.
Builds on
10 of 36 references.
Closest ancestors · 1
Same direction, newest first
Other references · 9
- 2025Qwen3 Technical Reportcited by 8,820
- 2025Qwen2.5-VL Technical Reportcited by 5,587
- 2024GPT-4o System Cardcited by 4,588
- 2023InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Taskscited by 2,565
- 2023SEED-Bench: Benchmarking Multimodal LLMs with Generative Comprehensioncited by 940
- 2024DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understandingcited by 572
- 2024CogVLM2: Visual Language Models for Image and Video Understandingcited by 217
- 2024CMMMU: A Chinese Massive Multi-discipline Multimodal Understanding Benchmarkcited by 56
- 2026VULCA-Bench: A Multicultural Vision-Language Benchmark for Evaluating Cultural Understandingcited by 3
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2025TCC-Bench: Benchmarking the Traditional Chinese Culture Understanding Capabilities of MLLMs
- 2025Seeing Culture: A Benchmark for Visual Reasoning and Grounding
- 2025Hanfu-Bench: A Multimodal Benchmark on Cross-Temporal Cultural Understanding and Transcreation
- 2024Benchmarking Vision Language Models for Cultural Understanding
- 2025VQArt-Bench: A semantically rich VQA Benchmark for Art and Cultural Heritage
- 2024CVQA: Culturally-diverse Multilingual Visual Question Answering Benchmark
- 2025Rice-VL: Evaluating Vision-Language Models for Cultural Understanding Across ASEAN Countries
- 2026CulMind: Benchmarking Multimodal Understanding and Reasoning in Chinese Cultural Heritage