EPE-P: Evidence-based Parameter-efficient Prompting for Multimodal Learning with Missing Modalities

TL;DR

EPE-P is a novel, efficient prompting method for multimodal learning that effectively handles missing modalities by reducing parameter redundancy and incorporating an evidence-based loss to improve robustness.

Contribution

The paper introduces EPE-P, a parameter-efficient prompting approach that unifies prompts across modalities and employs an evidence-based loss for better missing modality handling.

Findings

EPE-P outperforms existing methods in effectiveness.

EPE-P reduces model complexity and parameters.

EPE-P enhances robustness in multimodal learning with missing data.

Abstract

Missing modalities are a common challenge in real-world multimodal learning scenarios, occurring during both training and testing. Existing methods for managing missing modalities often require the design of separate prompts for each modality or missing case, leading to complex designs and a substantial increase in the number of parameters to be learned. As the number of modalities grows, these methods become increasingly inefficient due to parameter redundancy. To address these issues, we propose Evidence-based Parameter-Efficient Prompting (EPE-P), a novel and parameter-efficient method for pretrained multimodal networks. Our approach introduces a streamlined design that integrates prompting information across different modalities, reducing complexity and mitigating redundant parameters. Furthermore, we propose an Evidence-based Loss function to better handle the uncertainty associated with missing modalities, improving the model's decision-making. Our experiments demonstrate that EPE-P outperforms existing prompting-based methods in terms of both effectiveness and efficiency. The code is released at https://github.com/Boris-Jobs/EPE-P_MLLMs-Robustness.