IllusionVQA: A Challenging Optical Illusion Dataset for Vision Language Models

Haz Shahgir, Khondker S Sayeed, Abhik Bhattacharjee, Wasi Ahmad, Yue Dong, Rifat Shahriyar

Conference On Language Models 1 pre-proceedings (COLM 2024) Conference

BibTeX download is not available in the pre-proceeding

The advent of Vision Language Models (VLM) has allowed researchers to investigate the visual understanding of a neural network using natural language. Beyond object classification and detection, VLMs are capable of visual comprehension and common-sense reasoning. This naturally led to the question: How do VLMs respond when the image itself is inherently unreasonable? To this end, we present IllusionVQA: a diverse dataset of challenging optical illusions and hard-to-interpret scenes to test the capability of VLMs in two distinct multiple-choice VQA tasks - comprehension and soft localization. GPT4V, the best performing VLM, achieves 62.99\% accuracy (4-shot) on the comprehension task and 49.7\% on the localization task (4-shot and Chain-of-Thought). Human evaluation reveals that humans achieve 91.03\% and 100\% accuracy in comprehension and localization. We discover that In-Context Learning (ICL) and Chain-of-Thought reasoning substantially degrade the performance of Gemini-Pro on the localization task. Tangentially, we discover a potential weakness in the ICL capabilities of VLMs: they fail to locate optical illusions even when the correct answer is in the context window as a few-shot example.