LAMAE learns cross-lead interactions in ECGs via latent attention in a masked autoencoder, providing structural supervision that improves representation quality and outperforms baselines on ICD-10 code prediction.
Structure is supervision: Multiview masked autoencoders for radiology.arXiv preprint arXiv:2511.22294,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Foundation Model for Cardiac Time Series via Masked Latent Attention
LAMAE learns cross-lead interactions in ECGs via latent attention in a masked autoencoder, providing structural supervision that improves representation quality and outperforms baselines on ICD-10 code prediction.