TripletCLIP: Improving Compositional Reasoning of CLIP via Synthetic Vision-Language Negatives

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Patel, Maitreya, Kusumba, Abhiram, Cheng, Sheng, Kim, Changhoon, Gokhale, Tejas, Baral, Chitta, Yang, Yezhou
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866909377700560896
author Patel, Maitreya
Kusumba, Abhiram
Cheng, Sheng
Kim, Changhoon
Gokhale, Tejas
Baral, Chitta
Yang, Yezhou
author_facet Patel, Maitreya
Kusumba, Abhiram
Cheng, Sheng
Kim, Changhoon
Gokhale, Tejas
Baral, Chitta
Yang, Yezhou
contents Contrastive Language-Image Pretraining (CLIP) models maximize the mutual information between text and visual modalities to learn representations. This makes the nature of the training data a significant factor in the efficacy of CLIP for downstream tasks. However, the lack of compositional diversity in contemporary image-text datasets limits the compositional reasoning ability of CLIP. We show that generating ``hard'' negative captions via in-context learning and synthesizing corresponding negative images with text-to-image generators offers a solution. We introduce a novel contrastive pre-training strategy that leverages these hard negative captions and images in an alternating fashion to train CLIP. We demonstrate that our method, named TripletCLIP, when applied to existing datasets such as CC3M and CC12M, enhances the compositional capabilities of CLIP, resulting in an absolute improvement of over 9% on the SugarCrepe benchmark on an equal computational budget, as well as improvements in zero-shot image classification and image retrieval. Our code, models, and data are available at: https://tripletclip.github.io
format Preprint
id arxiv_https___arxiv_org_abs_2411_02545
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle TripletCLIP: Improving Compositional Reasoning of CLIP via Synthetic Vision-Language Negatives
Patel, Maitreya
Kusumba, Abhiram
Cheng, Sheng
Kim, Changhoon
Gokhale, Tejas
Baral, Chitta
Yang, Yezhou
Computer Vision and Pattern Recognition
Computation and Language
Contrastive Language-Image Pretraining (CLIP) models maximize the mutual information between text and visual modalities to learn representations. This makes the nature of the training data a significant factor in the efficacy of CLIP for downstream tasks. However, the lack of compositional diversity in contemporary image-text datasets limits the compositional reasoning ability of CLIP. We show that generating ``hard'' negative captions via in-context learning and synthesizing corresponding negative images with text-to-image generators offers a solution. We introduce a novel contrastive pre-training strategy that leverages these hard negative captions and images in an alternating fashion to train CLIP. We demonstrate that our method, named TripletCLIP, when applied to existing datasets such as CC3M and CC12M, enhances the compositional capabilities of CLIP, resulting in an absolute improvement of over 9% on the SugarCrepe benchmark on an equal computational budget, as well as improvements in zero-shot image classification and image retrieval. Our code, models, and data are available at: https://tripletclip.github.io
title TripletCLIP: Improving Compositional Reasoning of CLIP via Synthetic Vision-Language Negatives
topic Computer Vision and Pattern Recognition
Computation and Language
url https://arxiv.org/abs/2411.02545