Compositional Kronecker Context Optimization for Vision-Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ding, Kun, Li, Xiaohui, Yu, Qiang, Wang, Ying, Zhang, Haojian, Xiang, Shiming
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908315693350912
author Ding, Kun
Li, Xiaohui
Yu, Qiang
Wang, Ying
Zhang, Haojian
Xiang, Shiming
author_facet Ding, Kun
Li, Xiaohui
Yu, Qiang
Wang, Ying
Zhang, Haojian
Xiang, Shiming
contents Context Optimization (CoOp) has emerged as a simple yet effective technique for adapting CLIP-like vision-language models to downstream image recognition tasks. Nevertheless, learning compact context with satisfactory base-to-new, domain and cross-task generalization ability while adapting to new tasks is still a challenge. To tackle such a challenge, we propose a lightweight yet generalizable approach termed Compositional Kronecker Context Optimization (CK-CoOp). Technically, the prompt's context words in CK-CoOp are learnable vectors, which are crafted by linearly combining base vectors sourced from a dictionary. These base vectors consist of a non-learnable component obtained by quantizing the weights in the token embedding layer, and a learnable component constructed by applying Kronecker product on several learnable tiny matrices. Intuitively, the compositional structure mitigates the risk of overfitting on training data by remembering more pre-trained knowledge. Meantime, the Kronecker product breaks the non-learnable restrictions of the dictionary, thereby enhancing representation ability with minimal additional parameters. Extensive experiments confirm that CK-CoOp achieves state-of-the-art performance under base-to-new, domain and cross-task generalization evaluation, but also has the metrics of fewer learnable parameters and efficient training and inference speed.
format Preprint
id arxiv_https___arxiv_org_abs_2403_11631
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Compositional Kronecker Context Optimization for Vision-Language Models
Ding, Kun
Li, Xiaohui
Yu, Qiang
Wang, Ying
Zhang, Haojian
Xiang, Shiming
Computer Vision and Pattern Recognition
Context Optimization (CoOp) has emerged as a simple yet effective technique for adapting CLIP-like vision-language models to downstream image recognition tasks. Nevertheless, learning compact context with satisfactory base-to-new, domain and cross-task generalization ability while adapting to new tasks is still a challenge. To tackle such a challenge, we propose a lightweight yet generalizable approach termed Compositional Kronecker Context Optimization (CK-CoOp). Technically, the prompt's context words in CK-CoOp are learnable vectors, which are crafted by linearly combining base vectors sourced from a dictionary. These base vectors consist of a non-learnable component obtained by quantizing the weights in the token embedding layer, and a learnable component constructed by applying Kronecker product on several learnable tiny matrices. Intuitively, the compositional structure mitigates the risk of overfitting on training data by remembering more pre-trained knowledge. Meantime, the Kronecker product breaks the non-learnable restrictions of the dictionary, thereby enhancing representation ability with minimal additional parameters. Extensive experiments confirm that CK-CoOp achieves state-of-the-art performance under base-to-new, domain and cross-task generalization evaluation, but also has the metrics of fewer learnable parameters and efficient training and inference speed.
title Compositional Kronecker Context Optimization for Vision-Language Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2403.11631