EVE: Efficient Vision-Language Pre-training with Masked Prediction and Modality-Aware MoE

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Junyi, Guo, Longteng, Sun, Jia, Shao, Shuai, Yuan, Zehuan, Lin, Liang, Zhang, Dongyu
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917601313030144
author Chen, Junyi
Guo, Longteng
Sun, Jia
Shao, Shuai
Yuan, Zehuan
Lin, Liang
Zhang, Dongyu
author_facet Chen, Junyi
Guo, Longteng
Sun, Jia
Shao, Shuai
Yuan, Zehuan
Lin, Liang
Zhang, Dongyu
contents Building scalable vision-language models to learn from diverse, multimodal data remains an open challenge. In this paper, we introduce an Efficient Vision-languagE foundation model, namely EVE, which is one unified multimodal Transformer pre-trained solely by one unified pre-training task. Specifically, EVE encodes both vision and language within a shared Transformer network integrated with modality-aware sparse Mixture-of-Experts (MoE) modules, which capture modality-specific information by selectively switching to different experts. To unify pre-training tasks of vision and language, EVE performs masked signal modeling on image-text pairs to reconstruct masked signals, i.e., image pixels and text tokens, given visible signals. This simple yet effective pre-training objective accelerates training by 3.5x compared to the model pre-trained with Image-Text Contrastive and Image-Text Matching losses. Owing to the combination of the unified architecture and pre-training task, EVE is easy to scale up, enabling better downstream performance with fewer resources and faster training speed. Despite its simplicity, EVE achieves state-of-the-art performance on various vision-language downstream tasks, including visual question answering, visual reasoning, and image-text retrieval.
format Preprint
id arxiv_https___arxiv_org_abs_2308_11971
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle EVE: Efficient Vision-Language Pre-training with Masked Prediction and Modality-Aware MoE
Chen, Junyi
Guo, Longteng
Sun, Jia
Shao, Shuai
Yuan, Zehuan
Lin, Liang
Zhang, Dongyu
Computer Vision and Pattern Recognition
Computation and Language
Machine Learning
Multimedia
Building scalable vision-language models to learn from diverse, multimodal data remains an open challenge. In this paper, we introduce an Efficient Vision-languagE foundation model, namely EVE, which is one unified multimodal Transformer pre-trained solely by one unified pre-training task. Specifically, EVE encodes both vision and language within a shared Transformer network integrated with modality-aware sparse Mixture-of-Experts (MoE) modules, which capture modality-specific information by selectively switching to different experts. To unify pre-training tasks of vision and language, EVE performs masked signal modeling on image-text pairs to reconstruct masked signals, i.e., image pixels and text tokens, given visible signals. This simple yet effective pre-training objective accelerates training by 3.5x compared to the model pre-trained with Image-Text Contrastive and Image-Text Matching losses. Owing to the combination of the unified architecture and pre-training task, EVE is easy to scale up, enabling better downstream performance with fewer resources and faster training speed. Despite its simplicity, EVE achieves state-of-the-art performance on various vision-language downstream tasks, including visual question answering, visual reasoning, and image-text retrieval.
title EVE: Efficient Vision-Language Pre-training with Masked Prediction and Modality-Aware MoE
topic Computer Vision and Pattern Recognition
Computation and Language
Machine Learning
Multimedia
url https://arxiv.org/abs/2308.11971