Mar, 2024
IllusionVQA:一个为视觉语言模型设计的具有挑战性的视错觉数据集
IllusionVQA: A Challenging Optical Illusion Dataset for Vision Language Models
Haz Sameen Shahgir, Khondker Salman Sayeed, Abhik Bhattacharjee, Wasi Uddin Ahmad, Yue Dong...
TL;DRVision Language Models are tested on the IllusionVQA dataset, revealing their performance and weaknesses in comprehension and soft localization tasks, particularly in the context of optical illusions and In-Context Learning.