MCE: Towards a General Framework for Handling Missing Modalities under Imbalanced Missing Rates

Published in Pattern Recognition.

Modality Capability Enhancement (MCE) is a multi-modal learning framework designed to prevent performance drops when different data modalities are missing at imbalanced rates.

Key Features

Contains unimodal checkpoints (each modality), which supports:

  • Emotion recognition task on IEMOCAP dataset
  • Digit recognition task on AudiovisionMNIST dataset

Citation

If you use MCE and its checkpoints in your research, please use the following BibTeX entry.

@article{zhao2025mce,
  title={MCE: Towards a general framework for handling missing modalities under imbalanced missing rates},
  author={Zhao, Binyu and Zhang, Wei and Zou, Zhaonian},
  journal={Pattern Recognition},
  pages={112591},
  year={2025},
  publisher={Elsevier}
}

Links

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Paper for byzhaoAI/MCE_ckpts