mlsanityData to 14 Oct 2026

Benchmarking Attribute Discrimination in Infant-Scale Vision-Language Models

arXiv 2512.18951 ↗First posted 22 Dec 2025cs.LGcited by 0

On the map: Multimodal LLMs / vision-language models → MLLM training and visual instruction tuning

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in MLLM training and visual instruction tuning, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 7 of its 24 references that are in the map. First posted on arXiv 22 Dec 2025.

Builds on

7 of 24 references.

References · 7

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
  3. 2023Sigmoid Loss for Language Image Pre-Trainingcited by 3,109
  4. 2016Aggregated Residual Transformations for Deep Neural Networkscited by 2,546
  5. 2025ColorBench: Can VLMs See and Understand the Colorful World? A Comprehensive Benchmark for Color Perception, Reasoning, and Robustnesscited by 28
  6. 2022When are Lemons Purple? The Concept Association Bias of Vision-Language Modelscited by 23
  7. 2025Discovering Hidden Visual Concepts Beyond Linguistic Input in Infant Learningcited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025BabyVLM-V2: Toward Developmentally Grounded Pretraining and Benchmarking of Vision Foundation Models
  2. 2025BabyVLM: Data-Efficient Pretraining of VLMs Inspired by Infant Learning
  3. 2025Concepts Learned Visually by Infants Can Contribute to Visual Learning and Understanding in AI Models
  4. 2026NumerosityVLM: A Cognitively Inspired Benchmark for Interpreting Numerosity Representations in Vision-Language Models
  5. 2025Assessing the alignment between infants' visual and linguistic experience using multimodal language models
  6. 2026KidVis: Do Multimodal Large Language Models Possess the Visual Perceptual Capabilities of a 6-Year-Old?
  7. 2026BabyVision: Visual Reasoning Beyond Language
  8. 2024Discovering Object Attributes by Prompting Large Language Models with Perception-Action APIs