RA-BLIP: Multimodal Adaptive Retrieval-Augmented Bootstrapping Language-Image Pre-training

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ding, Muhe, Ma, Yang, Qin, Pengda, Wu, Jianlong, Li, Yuhong, Nie, Liqiang
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916444626747392
author Ding, Muhe
Ma, Yang
Qin, Pengda
Wu, Jianlong
Li, Yuhong
Nie, Liqiang
author_facet Ding, Muhe
Ma, Yang
Qin, Pengda
Wu, Jianlong
Li, Yuhong
Nie, Liqiang
contents Multimodal Large Language Models (MLLMs) have recently received substantial interest, which shows their emerging potential as general-purpose models for various vision-language tasks. MLLMs involve significant external knowledge within their parameters; however, it is challenging to continually update these models with the latest knowledge, which involves huge computational costs and poor interpretability. Retrieval augmentation techniques have proven to be effective plugins for both LLMs and MLLMs. In this study, we propose multimodal adaptive Retrieval-Augmented Bootstrapping Language-Image Pre-training (RA-BLIP), a novel retrieval-augmented framework for various MLLMs. Considering the redundant information within vision modality, we first leverage the question to instruct the extraction of visual information through interactions with one set of learnable queries, minimizing irrelevant interference during retrieval and generation. Besides, we introduce a pre-trained multimodal adaptive fusion module to achieve question text-to-multimodal retrieval and integration of multimodal knowledge by projecting visual and language modalities into a unified semantic space. Furthermore, we present an Adaptive Selection Knowledge Generation (ASKG) strategy to train the generator to autonomously discern the relevance of retrieved knowledge, which realizes excellent denoising performance. Extensive experiments on open multimodal question-answering datasets demonstrate that RA-BLIP achieves significant performance and surpasses the state-of-the-art retrieval-augmented models.
format Preprint
id arxiv_https___arxiv_org_abs_2410_14154
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle RA-BLIP: Multimodal Adaptive Retrieval-Augmented Bootstrapping Language-Image Pre-training
Ding, Muhe
Ma, Yang
Qin, Pengda
Wu, Jianlong
Li, Yuhong
Nie, Liqiang
Multimedia
Artificial Intelligence
Multimodal Large Language Models (MLLMs) have recently received substantial interest, which shows their emerging potential as general-purpose models for various vision-language tasks. MLLMs involve significant external knowledge within their parameters; however, it is challenging to continually update these models with the latest knowledge, which involves huge computational costs and poor interpretability. Retrieval augmentation techniques have proven to be effective plugins for both LLMs and MLLMs. In this study, we propose multimodal adaptive Retrieval-Augmented Bootstrapping Language-Image Pre-training (RA-BLIP), a novel retrieval-augmented framework for various MLLMs. Considering the redundant information within vision modality, we first leverage the question to instruct the extraction of visual information through interactions with one set of learnable queries, minimizing irrelevant interference during retrieval and generation. Besides, we introduce a pre-trained multimodal adaptive fusion module to achieve question text-to-multimodal retrieval and integration of multimodal knowledge by projecting visual and language modalities into a unified semantic space. Furthermore, we present an Adaptive Selection Knowledge Generation (ASKG) strategy to train the generator to autonomously discern the relevance of retrieved knowledge, which realizes excellent denoising performance. Extensive experiments on open multimodal question-answering datasets demonstrate that RA-BLIP achieves significant performance and surpasses the state-of-the-art retrieval-augmented models.
title RA-BLIP: Multimodal Adaptive Retrieval-Augmented Bootstrapping Language-Image Pre-training
topic Multimedia
Artificial Intelligence
url https://arxiv.org/abs/2410.14154