CMML uses a Cascade Residual Transformer Autoencoder with learnable context tokens and memory banks to synthesize missing modalities, followed by semantic alignment and contrastive refinement, achieving 1-1.3% AUC gains on three medical datasets.
IEEE Journal of Biomedical and Health Informatics
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Context-driven Missing-Modality Learning for Robust Medical Diagnosis with Image-Tabular Data
CMML uses a Cascade Residual Transformer Autoencoder with learnable context tokens and memory banks to synthesize missing modalities, followed by semantic alignment and contrastive refinement, achieving 1-1.3% AUC gains on three medical datasets.