BabyVision: Visual Reasoning Beyond Language

• URL: http://arxiv.org/abs/2601.06521v1
• Date: Sat, 10 Jan 2026 10:42:44 GMT
• Title: BabyVision: Visual Reasoning Beyond Language
• Authors: Liang Chen, Weichu Xie, Yiyan Liang, Hongfeng He, Hans Zhao, Zhibo Yang, Zhiqi Huang, Haoning Wu, Haoyu Lu, Y. charles, Yiping Bao, Yuantao Fan, Guopeng Li, Haiyang Shen, Xuanzhong Chen, Wendong Xu, Shuzheng Si, Zefan Cai, Wenhao Chai, Ziqi Huang, Fangfu Liu, Tianyu Liu, Baobao Chang, Xiaobo Hu, Kaiyuan Chen, Yixin Ren, Yang Liu, Yuan Gong, Kuan Li,
• Abstract summary: We introduce BabyVision, a benchmark designed to assess core visual abilities independent of linguistic knowledge.<n> Empirical results and human evaluation reveal that leading MLLMs perform significantly below human baselines.<n> Gemini3-Pro-Preview scores 49.7, lagging behind 6-year-old humans and falling well behind the average adult score of 94.1.
• Score: 60.43605497226761
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: While humans develop core visual skills long before acquiring language, contemporary Multimodal LLMs (MLLMs) still rely heavily on linguistic priors to compensate for their fragile visual understanding. We uncovered a crucial fact: state-of-the-art MLLMs consistently fail on basic visual tasks that humans, even 3-year-olds, can solve effortlessly. To systematically investigate this gap, we introduce BabyVision, a benchmark designed to assess core visual abilities independent of linguistic knowledge for MLLMs. BabyVision spans a wide range of tasks, with 388 items divided into 22 subclasses across four key categories. Empirical results and human evaluation reveal that leading MLLMs perform significantly below human baselines. Gemini3-Pro-Preview scores 49.7, lagging behind 6-year-old humans and falling well behind the average adult score of 94.1. These results show despite excelling in knowledge-heavy evaluations, current MLLMs still lack fundamental visual primitives. Progress in BabyVision represents a step toward human-level visual perception and reasoning capabilities. We also explore solving visual reasoning with generation models by proposing BabyVision-Gen and automatic evaluation toolkit. Our code and benchmark data are released at https://github.com/UniPat-AI/BabyVision for reproduction.

Related papers

• KidVis: Do Multimodal Large Language Models Possess the Visual Perceptual Capabilities of a 6-Year-Old? [79.27736230305516]
  We introduce KidVis, a novel benchmark grounded in the theory of human visual development.<n> evaluating 20 state-of-the-art MLLMs against a human physiological baseline reveals a stark performance disparity.<n>This study confirms that current MLLMs, despite their reasoning prowess, lack the essential physiological perceptual primitives required for generalized visual intelligence.
  arXiv  Detail & Related papers  (2026-01-13T07:32:50Z)
• VKnowU: Evaluating Visual Knowledge Understanding in Multimodal LLMs [35.79620808899466]
  Visual knowledge forms a bridge between perception and reasoning.<n> Evaluation of 23 SOTA MLLMs reveals that leading models still fall short of human performance.<n>We introduce a new dataset, VKnowQA, and VideoKnow+, a baseline model that explicitly incorporates visual knowledge into MLLMs.
  arXiv  Detail & Related papers  (2025-11-25T12:58:32Z)
• Pixels, Patterns, but No Poetry: To See The World like Humans [33.773551676022514]
  State-of-the-art MLLMs exhibit catastrophic failures on our perceptual tasks trivial for humans.<n>This paper shifts focus from reasoning to perception.
  arXiv  Detail & Related papers  (2025-07-21T21:50:16Z)
• VisuLogic: A Benchmark for Evaluating Visual Reasoning in Multi-modal Large Language Models [121.03333569013148]
  We introduce VisuLogic: a benchmark of 1,000 human-verified problems across six categories.<n>These types of questions can be evaluated to assess the visual reasoning capabilities of MLLMs from multiple perspectives.<n>Most models score below 30% accuracy-only slightly above the 25% random baseline and far below the 51.4% achieved by humans.
  arXiv  Detail & Related papers  (2025-04-21T17:59:53Z)
• Human Cognitive Benchmarks Reveal Foundational Visual Gaps in MLLMs [65.93003087656754]
  VisFactor is a benchmark that digitizes 20 vision-centric subtests from a well-established cognitive psychology assessment.<n>We evaluate 20 frontier Multimodal Large Language Models (MLLMs) from GPT, Gemini, Claude, LLaMA, Qwen, and SEED families.<n>The best-performing model achieves a score of only 25.19 out of 100, with consistent failures on tasks such as mental rotation, spatial relation inference, and figure-ground discrimination.
  arXiv  Detail & Related papers  (2025-02-23T04:21:32Z)
• Do Multimodal Large Language Models See Like Humans? [50.938168841711445]
  Multimodal Large Language Models (MLLMs) have achieved impressive results on various vision tasks, leveraging recent advancements in large language models.<n>Current benchmarks lack the ability to evaluate MLLMs from this perspective.<n>We introduce HVSBench, a large-scale benchmark designed to assess the alignment between MLLMs and the human visual system.
  arXiv  Detail & Related papers  (2024-12-12T18:59:25Z)
• Visualization Literacy of Multimodal Large Language Models: A Comparative Study [12.367399155606162]
  multimodal large language models (MLLMs) combine the inherent power of large language models (LLMs) with the renewed capabilities to reason about the multimodal context. Many recent works in visualization have demonstrated MLLMs' capability to understand and interpret visualization results and explain the content of the visualization to users in natural language. In this work, we aim to fill the gap by utilizing the concept of visualization literacy to evaluate MLLMs.
  arXiv  Detail & Related papers  (2024-06-24T17:52:16Z)
• What is the Visual Cognition Gap between Humans and Multimodal LLMs? [63.81347276258992]
  We evaluate the visual cognition capability of Multimodal Large Language Models (MLLMs) and compare their performance with human visual cognition studies.<n>Our comparative experiments with different baselines reveal a gap between MLLMs and human intelligence.<n>We believe that the public release of MaRs-VQA and the Qwen2-VCog baseline model will drive progress toward the next generation of MLLMs with human-like visual cognition abilities.
  arXiv  Detail & Related papers  (2024-06-14T22:02:21Z)
• Eyes Wide Shut? Exploring the Visual Shortcomings of Multimodal LLMs [50.77984109941538]
  Our research reveals that the visual capabilities in recent multimodal LLMs still exhibit systematic shortcomings. We identify ''CLIP-blind pairs'' - images that CLIP perceives as similar despite their clear visual differences. We evaluate various CLIP-based vision-and-language models and found a notable correlation between visual patterns that challenge CLIP models and those problematic for multimodal LLMs.
  arXiv  Detail & Related papers  (2024-01-11T18:58:36Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.