Dealing with All-stage Missing Modality: Towards A Universal Model with Robust Reconstruction and Personalization

TL;DR

This paper introduces a universal multimodal learning model capable of handling missing modalities during both training and testing by using reconstruction, co-distillation, and personalization techniques, validated on brain tumor segmentation tasks.

Contribution

The paper presents a novel universal model with robust reconstruction and personalization that effectively manages all-stage missing modalities in multimodal learning.

Findings

Outperforms previous methods on brain tumor segmentation benchmarks.

Effectively handles various missing modality ratios during training and testing.

Demonstrates robustness and adaptability across different missing data scenarios.

Abstract

Addressing missing modalities presents a critical challenge in multimodal learning. Current approaches focus on developing models that can handle modality-incomplete inputs during inference, assuming that the full set of modalities are available for all the data during training. This reliance on full-modality data for training limits the use of abundant modality-incomplete samples that are often encountered in practical settings. In this paper, we propose a robust universal model with modality reconstruction and model personalization, which can effectively tackle the missing modality at both training and testing stages. Our method leverages a multimodal masked autoencoder to reconstruct the missing modality and masked patches simultaneously, incorporating an innovative distribution approximation mechanism to fully utilize both modality-complete and modality-incomplete data. The reconstructed modalities then contributes to our designed data-model co-distillation scheme to guide the model learning in the presence of missing modalities. Moreover, we propose a CLIP-driven hyper-network to personalize partial model parameters, enabling the model to adapt to each distinct missing modality scenario. Our method has been extensively validated on two brain tumor segmentation benchmarks. Experimental results demonstrate the promising performance of our method, which consistently exceeds previous state-of-the-art approaches under the all-stage missing modality settings with different missing ratios. Code will be available.