Metraipolis › archive › dc_0d37e8ce205a
CulturalMenuBench: Probing the Knowledge-Application Gap in Multimodal Culinary Reasoning — arXiv:2609.03526
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.03526
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T01:10:49.552Z
- SHA-256 of the archived text
bc00d43d03d3b34caccc42c580a565b7de0919c4d505155fd392429b4ee7f755- Stable URL of this page
https://metraipolis.com/documento/dc_0d37e8ce205a
How to cite
CulturalMenuBench: Probing the Knowledge-Application Gap in Multimodal Culinary Reasoning — arXiv:2609.03526 — arXiv (en), https://arxiv.org/abs/2609.03526 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0d37e8ce205a (Metraipolis archive dc_0d37e8ce205a, sha256 bc00d43d03d3b34caccc42c580a565b7de0919c4d505155fd392429b4ee7f755)
BibTeX
@misc{dc0d37e8ce205a,
title = {CulturalMenuBench: Probing the Knowledge-Application Gap in Multimodal Culinary Reasoning — arXiv:2609.03526},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.03526},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0d37e8ce205a; sha256 bc00d43d03d3b34caccc42c580a565b7de0919c4d505155fd392429b4ee7f755},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.03526. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Multimodal language models achieve near-ceiling scores on food recognition benchmarks, yet it remains unclear whether this success reflects genuine cultural understanding or mere visual matching. To probe this distinction, we introduce CulturalMenuBench, a benchmark of 4,870 items in 10 languages across 18 regions; its 10 tasks pair final-dish and step-by-step cooking images with ingredients, procedural text, and regional labels, spanning basic recognition to process-grounded cultural attribution. Evaluating 12 models exposes a substantial knowledge-application gap: models exceeding 94% on standard multiple-choice tasks drop to at most 56% when attributing dishes to Chinese regional cuisines, despite an identical four-way format. Diagnostic analyses explain why: error patterns are consistent with random guessing, accuracy tracks visual distinctiveness rather than cultural structure, and models classify cuisines more accurately from dish names alone than from images (+7-18 points). The knowledge is thus present but cannot be activated through visual input. An ablation confirms these tasks genuinely require procedural evidence: removing sequential cooking images selectively degrades process-grounded tasks while others remain stable. Overall, CulturalMenuBench shows that near-perfect recognition can conceal an inability to apply cultural knowledge, motivating training that explicitly connects perception, procedure, and cultural context. Code and data are publicly available. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-04 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.03526 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-17T00:28:02.856Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-04?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).