{
  "annotation_id": "6f1b4343-4487-40e8-8e54-fc8385ea9e85",
  "date_created": "2026-02-17T05:53:15.044000Z",
  "date_modified": "2026-02-17T05:53:15.044000Z",
  "file_hash": "f78f3870955e09c9a7c004eda57a9d66e46d85749d7203fbb34bf169b1f7f733",
  "private": false,
  "record": {
    "abstract": "We introduce VULCA-Bench, a multicultural art-critique benchmark for evaluating Vision-Language Models\u0027 (VLMs) cultural understanding beyond surface-level visual perception. Existing VLM benchmarks predominantly measure L1-L2 capabilities (object recognition, scene description, and factual question answering) while under-evaluate higher-order cultural interpretation. VULCA-Bench contains 7,410 matched image-critique pairs spanning eight cultural traditions, with Chinese-English bilingual coverage. We operationalise cultural understanding using a five-layer framework (L1-L5, from Visual Perception to Philosophical Aesthetics), instantiated as 225 culture-specific dimensions and supported by expert-written bilingual critiques. Our pilot results indicate that higher-layer reasoning (L3-L5) is consistently more challenging than visual and technical analysis (L1-L2). The dataset, evaluation scripts, and annotation tools are available under CC BY 4.0 in the supplementary materials.",
    "arxiv_id": "2601.07986",
    "authors": [
      "Haorui Yu",
      "Ramon Ruiz-Dolz",
      "Diji Yang",
      "Hang He",
      "Fengrui Zhang",
      "Qiufeng Yi"
    ],
    "categories": [
      "cs.CL",
      "cs.CV"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "title": "VULCA-Bench: A Multicultural Vision-Language Benchmark for Evaluating Cultural Understanding",
    "url": "https://arxiv.org/abs/2601.07986",
    "version": "v1"
  },
  "schema_id": "dorsal/arxiv",
  "source": {
    "execution_id": "99f88aec-d6a5-451a-80e7-251cb47cc16d",
    "id": "arXiv Dataset",
    "type": "Model",
    "variant": "snapshot-2026-01-17",
    "version": "0.1.0"
  },
  "user_id": 1000002
}