Saved in:
Bibliographic Details
Main Authors: Lee, Gun, An, Subin, Baik, Sungyong, Lee, Soochahn
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2407.13808
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911961726320640
author Lee, Gun
An, Subin
Baik, Sungyong
Lee, Soochahn
author_facet Lee, Gun
An, Subin
Baik, Sungyong
Lee, Soochahn
contents We propose a novel prompt tuning method called CoAPT(Context Attribute words in Prompt Tuning) for few/zero-shot image classification. The core motivation is that attributes are descriptive words with rich information about a given concept. Thus, we aim to enrich text queries of existing prompt tuning methods, improving alignment between text and image embeddings in CLIP embedding space. To do so, CoAPT integrates attribute words as additional prompts within learnable prompt tuning and can be easily incorporated into various existing prompt tuning methods. To facilitate the incorporation of attributes into text embeddings and the alignment with image embeddings, soft prompts are trained together with an additional meta-network that generates input-image-wise feature biases from the concatenated feature encodings of the image-text combined queries. Our experiments demonstrate that CoAPT leads to considerable improvements for existing baseline methods on several few/zero-shot image classification tasks, including base-to-novel generalization, cross-dataset transfer, and domain generalization. Our findings highlight the importance of combining hard and soft prompts and pave the way for future research on the interplay between text and image latent spaces in pre-trained models.
format Preprint
id arxiv_https___arxiv_org_abs_2407_13808
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle CoAPT: Context Attribute words for Prompt Tuning
Lee, Gun
An, Subin
Baik, Sungyong
Lee, Soochahn
Computer Vision and Pattern Recognition
We propose a novel prompt tuning method called CoAPT(Context Attribute words in Prompt Tuning) for few/zero-shot image classification. The core motivation is that attributes are descriptive words with rich information about a given concept. Thus, we aim to enrich text queries of existing prompt tuning methods, improving alignment between text and image embeddings in CLIP embedding space. To do so, CoAPT integrates attribute words as additional prompts within learnable prompt tuning and can be easily incorporated into various existing prompt tuning methods. To facilitate the incorporation of attributes into text embeddings and the alignment with image embeddings, soft prompts are trained together with an additional meta-network that generates input-image-wise feature biases from the concatenated feature encodings of the image-text combined queries. Our experiments demonstrate that CoAPT leads to considerable improvements for existing baseline methods on several few/zero-shot image classification tasks, including base-to-novel generalization, cross-dataset transfer, and domain generalization. Our findings highlight the importance of combining hard and soft prompts and pave the way for future research on the interplay between text and image latent spaces in pre-trained models.
title CoAPT: Context Attribute words for Prompt Tuning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2407.13808