Recently, Large Language Models (LLMs) have been serving as general-purpose interfaces, posing a significant demand for comprehensive visual knowledge. However, it remains unclear how well current LLMs and their visually augmented counterparts (VaLMs) can master visual commonsense knowledge. To investigate this, we propose ImageNetVC, a human-annotated dataset specifically designed for zero- and few-shot visual commonsense evaluation across 1,000 ImageNet categories. Utilizing ImageNetVC, we benchmark the fundamental visual commonsense knowledge of both unimodal LLMs and VaLMs. Furthermore, we analyze the factors affecting the visual commonsense knowledge of large-scale models, providing insights into the development of language models enriched with visual commonsense knowledge. Our code and dataset are available at https://github.com/hemingkx/ImageNetVC.
@article{arxiv.2305.15028,
title = {ImageNetVC: Zero- and Few-Shot Visual Commonsense Evaluation on 1000 ImageNet Categories},
author = {Heming Xia and Qingxiu Dong and Lei Li and Jingjing Xu and Tianyu Liu and Ziwei Qin and Zhifang Sui},
journal= {arXiv preprint arXiv:2305.15028},
year = {2023}
}
Comments
EMNLP 2023 Findings (Long Paper), camera-ready version