Embracing Language Inclusivity and Diversity in CLIP through Continual Language Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Bang, Dai, Yong, Cheng, Xuxin, Li, Yaowei, Raza, Asif, Zou, Yuexian
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913215971065856
author Yang, Bang
Dai, Yong
Cheng, Xuxin
Li, Yaowei
Raza, Asif
Zou, Yuexian
author_facet Yang, Bang
Dai, Yong
Cheng, Xuxin
Li, Yaowei
Raza, Asif
Zou, Yuexian
contents While vision-language pre-trained models (VL-PTMs) have advanced multimodal research in recent years, their mastery in a few languages like English restricts their applicability in broader communities. To this end, there is an increasing interest in developing multilingual VL models via a joint-learning setup, which, however, could be unrealistic due to expensive costs and data availability. In this work, we propose to extend VL-PTMs' language capacity by continual language learning (CLL), where a model needs to update its linguistic knowledge incrementally without suffering from catastrophic forgetting (CF). We begin our study by introducing a model dubbed CLL-CLIP, which builds upon CLIP, a prevailing VL-PTM that has acquired image-English text alignment. Specifically, CLL-CLIP contains an expandable token embedding layer to handle linguistic differences. It solely trains token embeddings to improve memory stability and is optimized under cross-modal and cross-lingual objectives to learn the alignment between images and multilingual texts. To alleviate CF raised by covariate shift and lexical overlap, we further propose a novel approach that ensures the identical distribution of all token embeddings during initialization and regularizes token embedding learning during training. We construct a CLL benchmark covering 36 languages based on MSCOCO and XM3600 datasets and then evaluate multilingual image-text retrieval performance. Extensive experiments verify the effectiveness of CLL-CLIP and show that our approach can boost CLL-CLIP, e.g., by 6.7% in text-to-image average Recall@1 on XM3600, and improve various state-of-the-art methods consistently. Our code and data are available at \url{https://github.com/yangbang18/CLFM}.
format Preprint
id arxiv_https___arxiv_org_abs_2401_17186
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Embracing Language Inclusivity and Diversity in CLIP through Continual Language Learning
Yang, Bang
Dai, Yong
Cheng, Xuxin
Li, Yaowei
Raza, Asif
Zou, Yuexian
Computer Vision and Pattern Recognition
Artificial Intelligence
Information Retrieval
While vision-language pre-trained models (VL-PTMs) have advanced multimodal research in recent years, their mastery in a few languages like English restricts their applicability in broader communities. To this end, there is an increasing interest in developing multilingual VL models via a joint-learning setup, which, however, could be unrealistic due to expensive costs and data availability. In this work, we propose to extend VL-PTMs' language capacity by continual language learning (CLL), where a model needs to update its linguistic knowledge incrementally without suffering from catastrophic forgetting (CF). We begin our study by introducing a model dubbed CLL-CLIP, which builds upon CLIP, a prevailing VL-PTM that has acquired image-English text alignment. Specifically, CLL-CLIP contains an expandable token embedding layer to handle linguistic differences. It solely trains token embeddings to improve memory stability and is optimized under cross-modal and cross-lingual objectives to learn the alignment between images and multilingual texts. To alleviate CF raised by covariate shift and lexical overlap, we further propose a novel approach that ensures the identical distribution of all token embeddings during initialization and regularizes token embedding learning during training. We construct a CLL benchmark covering 36 languages based on MSCOCO and XM3600 datasets and then evaluate multilingual image-text retrieval performance. Extensive experiments verify the effectiveness of CLL-CLIP and show that our approach can boost CLL-CLIP, e.g., by 6.7% in text-to-image average Recall@1 on XM3600, and improve various state-of-the-art methods consistently. Our code and data are available at \url{https://github.com/yangbang18/CLFM}.
title Embracing Language Inclusivity and Diversity in CLIP through Continual Language Learning
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Information Retrieval
url https://arxiv.org/abs/2401.17186