ImageNetVC: Zero- and Few-Shot Visual Commonsense Evaluation on 1000 ImageNet Categories
This work addresses the need for comprehensive visual knowledge in AI systems, particularly for researchers and developers working on multimodal models, though it is incremental as it focuses on evaluation rather than novel model development.
The authors tackled the problem of evaluating visual commonsense knowledge in large language models and their visually augmented counterparts by introducing ImageNetVC, a human-annotated dataset for zero- and few-shot evaluation across 1,000 ImageNet categories, and used it to benchmark these models and analyze factors affecting their performance.
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.