Decomposing Complex Visual Comprehension into Atomic Visual Skills for Vision Language Models

• URL: http://arxiv.org/abs/2505.20021v1
• Date: Mon, 26 May 2025 14:09:24 GMT
• Title: Decomposing Complex Visual Comprehension into Atomic Visual Skills for Vision Language Models
• Authors: Hyunsik Chae, Seungwoo Yoon, Jaden Park, Chloe Yewon Chun, Yongin Cho, Mu Cai, Yong Jae Lee, Ernest K. Ryu,
• Abstract summary: Recent Vision-Language Models (VLMs) have demonstrated impressive multimodal comprehension and reasoning capabilities.<n>We systematically categorize the fundamental, indivisible visual perception skills, which we refer to as atomic visual skills.<n>We benchmark state-of-the-art VLMs and find that they struggle with these tasks, despite being trivial for adult humans.
• Score: 28.88723703297274
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent Vision-Language Models (VLMs) have demonstrated impressive multimodal comprehension and reasoning capabilities, yet they often struggle with trivially simple visual tasks. In this work, we focus on the domain of basic 2D Euclidean geometry and systematically categorize the fundamental, indivisible visual perception skills, which we refer to as atomic visual skills. We then introduce the Atomic Visual Skills Dataset (AVSD) for evaluating VLMs on the atomic visual skills. Using AVSD, we benchmark state-of-the-art VLMs and find that they struggle with these tasks, despite being trivial for adult humans. Our findings highlight the need for purpose-built datasets to train and evaluate VLMs on atomic, rather than composite, visual perception tasks.

Related papers

• Response Wide Shut? Surprising Observations in Basic Vision Language Model Capabilities [54.94982467313341]
  Vision-language Models (VLMs) have emerged as general-purpose tools for addressing a variety of complex computer vision problems.<n>We set out to understand the limitations of SoTA VLMs on fundamental visual tasks by constructing a series of tests that probe which components of design, specifically, may be lacking.
  arXiv  Detail & Related papers  (2025-07-10T15:26:41Z)
• ViCrit: A Verifiable Reinforcement Learning Proxy Task for Visual Perception in VLMs [98.27348724529257]
  We introduce ViCrit (Visual Caption Hallucination Critic), an RL proxy task that trains VLMs to localize a subtle, synthetic visual hallucination injected into paragraphs of human-written image captions.<n>Models trained with the ViCrit Task exhibit substantial gains across a variety of vision-language models benchmarks.
  arXiv  Detail & Related papers  (2025-06-11T19:16:54Z)
• Hidden in plain sight: VLMs overlook their visual representations [48.83628674170634]
  We compare vision language models (VLMs) to their visual encoders to understand their ability to integrate across these modalities.<n>We find that VLMs perform substantially worse than their visual encoders, dropping to near-chance performance.
  arXiv  Detail & Related papers  (2025-06-09T17:59:54Z)
• Visual Language Models show widespread visual deficits on neuropsychological tests [0.0]
  We use the toolkit of neuropsychology to assess the capabilities of three state-of-the-art Visual Language Models (VLMs)<n>We find widespread deficits in low- and mid-level visual abilities that would be considered clinically significant in humans.<n>These selective deficits, profiled through validated test batteries, suggest that an artificial system can achieve complex object recognition without developing foundational visual concepts that in humans require no explicit training.
  arXiv  Detail & Related papers  (2025-04-15T01:04:56Z)
• Do we Really Need Visual Instructions? Towards Visual Instruction-Free Fine-tuning for Large Vision-Language Models [127.38740043393527]
  We propose ViFT, a visual instruction-free fine-tuning framework for LVLMs.<n>We only require the text-only instructions and image caption data during training, to separately learn the task-solving and visual perception abilities.<n> Experimental results demonstrate that ViFT can achieve state-of-the-art performance on several visual reasoning and visual instruction following benchmarks.
  arXiv  Detail & Related papers  (2025-02-17T04:38:12Z)
• COSMOS: Cross-Modality Self-Distillation for Vision Language Pre-training [49.2684130383925]
  We propose COSMOS: CrOSs-MOdality Self-distillation for vision-language pre-training.<n>It integrates a novel text-cropping strategy and cross-attention module into a self-supervised learning framework.<n>It consistently outperforms previous strong baselines on various zero-shot downstream tasks.
  arXiv  Detail & Related papers  (2024-12-02T18:56:06Z)
• VL-GLUE: A Suite of Fundamental yet Challenging Visuo-Linguistic Reasoning Tasks [48.67062958311173]
  VL-GLUE is a multitask benchmark for natural language understanding. We show that this benchmark is quite challenging for existing large-scale vision-language models.
  arXiv  Detail & Related papers  (2024-10-17T15:27:17Z)
• TWIST & SCOUT: Grounding Multimodal LLM-Experts by Forget-Free Tuning [54.033346088090674]
  We introduce TWIST & SCOUT, a framework that equips pre-trained MLLMs with visual grounding ability.<n>To fine-tune the model effectively, we generate a high-quality synthetic dataset we call SCOUT.<n>This dataset provides rich supervision signals, describing a step-by-step multimodal reasoning process.
  arXiv  Detail & Related papers  (2024-10-14T13:35:47Z)
• Response Wide Shut: Surprising Observations in Basic Vision Language Model Capabilities [30.176918208200604]
  Vision-Language Models (VLMs) have emerged as general purpose tools for addressing a variety of complex computer vision problems. These models have been shown to be highly capable, but also lacking some basic visual understanding skills. This paper sets out to understand the limitations of SoTA VLMs on fundamental visual tasks.
  arXiv  Detail & Related papers  (2024-08-13T08:26:32Z)
• MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contexts [170.01089233942594]
  MathVista is a benchmark designed to combine challenges from diverse mathematical and visual tasks. The best-performing GPT-4V model achieves an overall accuracy of 49.9%, substantially outperforming Bard, the second-best performer, by 15.1%. GPT-4V still falls short of human performance by 10.4%, as it often struggles to understand complex figures and perform rigorous reasoning.
  arXiv  Detail & Related papers  (2023-10-03T17:57:24Z)
• VIPHY: Probing "Visible" Physical Commonsense Knowledge [22.00069189468524]
  Vision-language models (VLMs) have shown remarkable performance on visual reasoning tasks. We evaluate their ability to acquire "visible" physical knowledge. Our results indicate a severe gap between model and human performance.
  arXiv  Detail & Related papers  (2022-09-15T02:06:25Z)

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.