Retrieval-Augmented Dynamic Prompt Tuning for Incomplete Multimodal Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lang, Jian, Cheng, Zhangtao, Zhong, Ting, Zhou, Fan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908407560142848
author Lang, Jian
Cheng, Zhangtao
Zhong, Ting
Zhou, Fan
author_facet Lang, Jian
Cheng, Zhangtao
Zhong, Ting
Zhou, Fan
contents Multimodal learning with incomplete modality is practical and challenging. Recently, researchers have focused on enhancing the robustness of pre-trained MultiModal Transformers (MMTs) under missing modality conditions by applying learnable prompts. However, these prompt-based methods face several limitations: (1) incomplete modalities provide restricted modal cues for task-specific inference, (2) dummy imputation for missing content causes information loss and introduces noise, and (3) static prompts are instance-agnostic, offering limited knowledge for instances with various missing conditions. To address these issues, we propose RAGPT, a novel Retrieval-AuGmented dynamic Prompt Tuning framework. RAGPT comprises three modules: (I) the multi-channel retriever, which identifies similar instances through a within-modality retrieval strategy, (II) the missing modality generator, which recovers missing information using retrieved contexts, and (III) the context-aware prompter, which captures contextual knowledge from relevant instances and generates dynamic prompts to largely enhance the MMT's robustness. Extensive experiments conducted on three real-world datasets show that RAGPT consistently outperforms all competitive baselines in handling incomplete modality problems. The code of our work and prompt-based baselines is available at https://github.com/Jian-Lang/RAGPT.
format Preprint
id arxiv_https___arxiv_org_abs_2501_01120
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Retrieval-Augmented Dynamic Prompt Tuning for Incomplete Multimodal Learning
Lang, Jian
Cheng, Zhangtao
Zhong, Ting
Zhou, Fan
Computer Vision and Pattern Recognition
Artificial Intelligence
Multimodal learning with incomplete modality is practical and challenging. Recently, researchers have focused on enhancing the robustness of pre-trained MultiModal Transformers (MMTs) under missing modality conditions by applying learnable prompts. However, these prompt-based methods face several limitations: (1) incomplete modalities provide restricted modal cues for task-specific inference, (2) dummy imputation for missing content causes information loss and introduces noise, and (3) static prompts are instance-agnostic, offering limited knowledge for instances with various missing conditions. To address these issues, we propose RAGPT, a novel Retrieval-AuGmented dynamic Prompt Tuning framework. RAGPT comprises three modules: (I) the multi-channel retriever, which identifies similar instances through a within-modality retrieval strategy, (II) the missing modality generator, which recovers missing information using retrieved contexts, and (III) the context-aware prompter, which captures contextual knowledge from relevant instances and generates dynamic prompts to largely enhance the MMT's robustness. Extensive experiments conducted on three real-world datasets show that RAGPT consistently outperforms all competitive baselines in handling incomplete modality problems. The code of our work and prompt-based baselines is available at https://github.com/Jian-Lang/RAGPT.
title Retrieval-Augmented Dynamic Prompt Tuning for Incomplete Multimodal Learning
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2501.01120