Robust Multimodal Learning with Missing Modalities via Parameter-Efficient Adaptation

TL;DR

This paper introduces a parameter-efficient adaptation method for pretrained multimodal networks that enhances robustness to missing modalities by modulating intermediate features, outperforming existing approaches across diverse tasks and datasets.

Contribution

The authors propose a simple, parameter-efficient adaptation technique that improves robustness of multimodal models to missing data by modulating intermediate features, requiring less than 1% additional parameters.

Findings

Improves performance with missing modalities across five tasks.

Requires fewer than 1% of total parameters for adaptation.

Outperforms existing methods in robustness to missing modalities.

Abstract

Multimodal learning seeks to utilize data from multiple sources to improve the overall performance of downstream tasks. It is desirable for redundancies in the data to make multimodal systems robust to missing or corrupted observations in some correlated modalities. However, we observe that the performance of several existing multimodal networks significantly deteriorates if one or multiple modalities are absent at test time. To enable robustness to missing modalities, we propose a simple and parameter-efficient adaptation procedure for pretrained multimodal networks. In particular, we exploit modulation of intermediate features to compensate for the missing modalities. We demonstrate that such adaptation can partially bridge performance drop due to missing modalities and outperform independent, dedicated networks trained for the available modality combinations in some cases. The proposed adaptation requires extremely small number of parameters (e.g., fewer than 1% of the total parameters) and applicable to a wide range of modality combinations and tasks. We conduct a series of experiments to highlight the missing modality robustness of our proposed method on five different multimodal tasks across seven datasets. Our proposed method demonstrates versatility across various tasks and datasets, and outperforms existing methods for robust multimodal learning with missing modalities.