Visual In-Context Learning for Large Vision-Language Models

Yucheng Zhou, Xiang Li, Qianning Wang, Jianbing Shen


Abstract
In Large Visual Language Models (LVLMs), the efficacy of In-Context Learning (ICL) remains limited by challenges in cross-modal interactions and representation disparities. To overcome these challenges, we introduce a novel Visual In-Context Learning (VICL) method comprising Visual Demonstration Retrieval, Intent-Oriented Image Summarization, and Intent-Oriented Demonstration Composition. Our approach retrieves images via ”Retrieval & Rerank” paradigm, summarises images with task intent and task-specific visual parsing, and composes language-based demonstrations that reduce token count and alleviate cross-modal interaction problem. Experimental evaluations on five visual reasoning datasets demonstrate the effectiveness of our method. Moreover, our extensive experiments leverage information flow analysis to elucidate the effectiveness of our method, and investigate the impact of length and position of demonstrations for LVLM. The use of in-context unlearning further shows promise in resetting specific model knowledge without retraining.
Anthology ID:
2024.findings-acl.940
Volume:
Findings of the Association for Computational Linguistics: ACL 2024
Month:
August
Year:
2024
Address:
Bangkok, Thailand
Editors:
Lun-Wei Ku, Andre Martins, Vivek Srikumar
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
15890–15902
Language:
URL:
https://aclanthology.org/2024.findings-acl.940
DOI:
10.18653/v1/2024.findings-acl.940
Bibkey:
Cite (ACL):
Yucheng Zhou, Xiang Li, Qianning Wang, and Jianbing Shen. 2024. Visual In-Context Learning for Large Vision-Language Models. In Findings of the Association for Computational Linguistics: ACL 2024, pages 15890–15902, Bangkok, Thailand. Association for Computational Linguistics.
Cite (Informal):
Visual In-Context Learning for Large Vision-Language Models (Zhou et al., Findings 2024)
Copy Citation:
PDF:
https://preview.aclanthology.org/autopr/2024.findings-acl.940.pdf