Answers are grounded in this paper and cite their sources.
Reveals aesthetic score disparities in vision-language models for East Asian art, suggesting processing asymmetries.
Vision–language models (VLMs) increasingly evaluate visual content, yet their behaviour across cultural traditions remains poorly characterised. We show that two open-weight VLMs, Qwen3-VL-8B and Llama-3.2-11B-Vision, assign systematically lower aesthetic-quality scores (model-generated ratings on a 1–10 scale) to East Asian art than to Western art, with Cohen’s d = \(-0.46\) and \(-0.36\) ( \(p < 10⁻¹⁶\) ). Internal probing reveals this disparity is accompanied by higher encoding cost under the primary reference sentence, elevated perplexity, reduced token confidence, and increased hidden-state norms. Across the two tested models, fifteen of eighteen signals shift in the same direction, indicating a consistent cross-model pattern; Llama cross-attention entropy shows a large effect ( d = \(+1.30\) , \(p < 10⁻¹⁷⁸\) ). The gap persists across decoding temperatures ( \(T ∈ \{0,\, 0.5,\, 1\}\) ; all \(p < 10⁻⁷\) ), reproduces at the group level under repeated stochastic sampling, and remains when the rating prompt’s cultural-significance criterion is removed, suggesting that it is not driven solely by stochastic decoding or by an explicit cultural-evaluation instruction. Hidden-state clustering places East Asian art in broadly overlapping representational regions relative to Western art, and keyword-based response analysis shows frequent Chinese lexical attribution for Korean artworks (88% Qwen, 65% Llama). In Qwen, matched-complexity analysis shows 87% of the score gap persists after controlling for image spectral properties. These results show that internal probing can help detect cross-cultural processing differences in the tested VLMs.
No takes yet. Share an insight, caveat, or question.
Bae et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: