Score: 2

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

Published: May 26, 2025 | arXiv ID: 2505.20021v1

By: Hyunsik Chae , Seungwoo Yoon , Jaden Park and more

Potential Business Impact:

Teaches computers to see basic shapes like humans.

Business Areas:
Image Recognition Data and Analytics, Software

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.


Page Count
69 pages

Category
Computer Science:
CV and Pattern Recognition