AAPL: Adding Attributes to Prompt Learning for Vision-Language Models

TL;DR

This paper introduces AAPL, a novel prompt learning method that adds attributes to improve vision-language models' ability to generalize to unseen classes, especially in zero-shot and few-shot tasks.

Contribution

The paper proposes adversarial token embedding and a new attribute-adding mechanism to enhance prompt learning for better unseen class generalization in vision-language models.

Findings

AAPL outperforms existing methods in zero-shot and few-shot learning.

AAPL demonstrates strong cross-dataset and domain generalization performance.

The method effectively disentangles visual bias from class features.

Abstract

Recent advances in large pre-trained vision-language models have demonstrated remarkable performance on zero-shot downstream tasks. Building upon this, recent studies, such as CoOp and CoCoOp, have proposed the use of prompt learning, where context within a prompt is replaced with learnable vectors, leading to significant improvements over manually crafted prompts. However, the performance improvement for unseen classes is still marginal, and to tackle this problem, data augmentation has been frequently used in traditional zero-shot learning techniques. Through our experiments, we have identified important issues in CoOp and CoCoOp: the context learned through traditional image augmentation is biased toward seen classes, negatively impacting generalization to unseen classes. To address this problem, we propose adversarial token embedding to disentangle low-level visual augmentation features from high-level class information when inducing bias in learnable prompts. Through our novel mechanism called "Adding Attributes to Prompt Learning", AAPL, we guide the learnable context to effectively extract text features by focusing on high-level features for unseen classes. We have conducted experiments across 11 datasets, and overall, AAPL shows favorable performances compared to the existing methods in few-shot learning, zero-shot learning, cross-dataset, and domain generalization tasks.