{"ID":2826036,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.18951","arxiv_id":"2512.18951","title":"Benchmarking Attribute Discrimination in Infant-Scale Vision-Language Models","abstract":"Infants learn not only object categories but also fine-grained visual attributes such as color, size, and texture from limited experience. Prior infant-scale vision--language models have mainly been evaluated on object recognition, leaving open whether they support within-class attribute discrimination. We introduce a controlled benchmark that varies color, size, and texture across 67 everyday object classes using synthetic rendering to decouple attribute values from object identity. We evaluate infant-trained models (CVCL and an infant-trained DINO baseline) against web-scale and ImageNet models (CLIP, SigLIP, ResNeXt) under two complementary settings: an image-only prototype test and a text--vision test with attribute--object prompts. We find a dissociation between visual and linguistic attribute information: infant-trained models form strong visual representations for size and discriminate texture comparably to other models, but perform poorly on visual color discrimination, and in the text--vision setting they struggle to ground color and show only modest size grounding. In contrast, web-trained vision--language models strongly ground color from text while exhibiting weaker visual size discrimination.","short_abstract":"Infants learn not only object categories but also fine-grained visual attributes such as color, size, and texture from limited experience. Prior infant-scale vision--language models have mainly been evaluated on object recognition, leaving open whether they support within-class attribute discrimination. We introduce a...","url_abs":"https://arxiv.org/abs/2512.18951","url_pdf":"https://arxiv.org/pdf/2512.18951v3","authors":"[\"Patrick Batsell\",\"Satoshi Tsutsui\",\"Bihan Wen\"]","published":"2025-12-22T01:58:17Z","proceeding":"cs.LG","tasks":"[\"cs.LG\"]","methods":"[\"Language Model\"]","has_code":false}