Understanding Retrieval-Augmented Task Adaptation for Vision-Language Models

arXiv:2405.01468v1 Announce Type: new
Abstract: Pre-trained contrastive vision-language models have demonstrated remarkable performance across a wide range of tasks. However, they often struggle on fine-trained datasets with categories not adequately represented during pre-training, which makes adaptation necessary. Recent works have shown promising results by utilizing samples from web-scale databases for retrieval-augmented adaptation, especially in low-data regimes. Despite the empirical success, understanding how retrieval impacts the adaptation of vision-language models remains an open research question. In this work, we adopt a …

abstract arxiv cs.ai cs.cv cs.lg databases datasets however language language models performance pre-training results retrieval retrieval-augmented samples scale struggle tasks training type understanding vision vision-language vision-language models web