ImageNetVC: Zero- and Few-Shot Visual Commonsense Evaluation on 1000 ImageNet Categories

Heming Xia, Qingxiu Dong, Lei Li, Jingjing Xu, Tianyu Liu +2 more
5/24/2023
cs.CLcs.AIcs.CV

Official code: hemingkx/ImageNetVC (10 stars, last commit 2023-12-09).

Abstract

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.

View on arXivView PDF

Code Implementations(1)

hemingkx/ImageNetVCOfficial100%

Codes and datasets for our paper "ImageNetVC: Zero- and Few-Shot Visual Commonsense Evaluation on 1000 ImageNet Categories" (EMNLP 2023 Findings).

102PythonMay 23, 20232 years ago

Similar Papers

VCD: A Dataset for Visual Commonsense Discovery in Images

92% match2024

ViCor: Bridging Visual Understanding and Commonsense Reasoning with Large Language Models

88% match1 repo42023

DIVE: Towards Descriptive and Diverse Visual Commonsense Generation

86% match2024

CommonsenseVIS: Visualizing and Understanding Commonsense Reasoning Capabilities of Natural Language Models

86% match2023

Gemini in Reasoning: Unveiling Commonsense in Multimodal Large Language Models

86% match1 repo382023

Improving Visual Commonsense in Language Models via Multiple Image Generation

85% match2024

Cite this paper

@article{xia2023imagenetvc,
  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},
  year   = {2023},
  eprint = {2305.15028},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2305.15028v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2305.15028.svg)](https://codeofpaper.com/paper/2305.15028)

Discussion