TransAgent: Transfer Vision-Language Foundation Models with Heterogeneous Agent Collaboration

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Guo, Yiwei, Zhuang, Shaobin, Li, Kunchang, Qiao, Yu, Wang, Yali
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917834307665920
author Guo, Yiwei
Zhuang, Shaobin
Li, Kunchang
Qiao, Yu
Wang, Yali
author_facet Guo, Yiwei
Zhuang, Shaobin
Li, Kunchang
Qiao, Yu
Wang, Yali
contents Vision-language foundation models (such as CLIP) have recently shown their power in transfer learning, owing to large-scale image-text pre-training. However, target domain data in the downstream tasks can be highly different from the pre-training phase, which makes it hard for such a single model to generalize well. Alternatively, there exists a wide range of expert models that contain diversified vision and/or language knowledge pre-trained on different modalities, tasks, networks, and datasets. Unfortunately, these models are "isolated agents" with heterogeneous structures, and how to integrate their knowledge for generalizing CLIP-like models has not been fully explored. To bridge this gap, we propose a general and concise TransAgent framework, which transports the knowledge of the isolated agents in a unified manner, and effectively guides CLIP to generalize with multi-source knowledge distillation. With such a distinct framework, we flexibly collaborate with 11 heterogeneous agents to empower vision-language foundation models, without further cost in the inference phase. Finally, our TransAgent achieves state-of-the-art performance on 11 visual recognition datasets. Under the same low-shot setting, it outperforms the popular CoOp with around 10% on average, and 20% on EuroSAT which contains large domain shifts.
format Preprint
id arxiv_https___arxiv_org_abs_2410_12183
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle TransAgent: Transfer Vision-Language Foundation Models with Heterogeneous Agent Collaboration
Guo, Yiwei
Zhuang, Shaobin
Li, Kunchang
Qiao, Yu
Wang, Yali
Computer Vision and Pattern Recognition
Vision-language foundation models (such as CLIP) have recently shown their power in transfer learning, owing to large-scale image-text pre-training. However, target domain data in the downstream tasks can be highly different from the pre-training phase, which makes it hard for such a single model to generalize well. Alternatively, there exists a wide range of expert models that contain diversified vision and/or language knowledge pre-trained on different modalities, tasks, networks, and datasets. Unfortunately, these models are "isolated agents" with heterogeneous structures, and how to integrate their knowledge for generalizing CLIP-like models has not been fully explored. To bridge this gap, we propose a general and concise TransAgent framework, which transports the knowledge of the isolated agents in a unified manner, and effectively guides CLIP to generalize with multi-source knowledge distillation. With such a distinct framework, we flexibly collaborate with 11 heterogeneous agents to empower vision-language foundation models, without further cost in the inference phase. Finally, our TransAgent achieves state-of-the-art performance on 11 visual recognition datasets. Under the same low-shot setting, it outperforms the popular CoOp with around 10% on average, and 20% on EuroSAT which contains large domain shifts.
title TransAgent: Transfer Vision-Language Foundation Models with Heterogeneous Agent Collaboration
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2410.12183