AIIC AI Intelligence Centre

SOURCE-LINKED INTELLIGENCE

Temporal Island Sparse Autoencoders for Interpreting Clinical Time-Series Models

arXiv · AI, language, vision and robotics · article · Nov 2, 2026 · UTC

Deep learning models for electronic health records achieve strong predictive performance but provide little insight into what clinical concepts they represent internally. Post-hoc attribution methods produce per-cell importance scores that lack temporal coherence, cross-variable structure, and reusability across patients. We introduce the Temporal Island Sparse Autoencoder (TI-SAE), which learns a backbone-specific, patient-shared clinical concept dictionary from a frozen predictive model's representations. Each dictionary entry (latent) is defined by learned temporal Gaussian islands (specify

Read original source ↗ Open in workspace

recordType
paper
region
Global

Evidence & attribution

First collected: 2026-11-03T08:23:14.856Z. This is not the publication date.