Sparse autoencoders applied to a 14.5M-parameter clinical EHR model reveal progressive abstraction across layers, with SAE features outperforming dense ones for mortality in full-sequence probes but not in leakage-safe windows where dense representations match or exceed them.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.LG 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
FlatASCEND generates conditional clinical event sequences that partially recover known mechanistic drug associations from observational data but fail to maintain them under direct preference optimization and show weaker performance on longer outpatient timelines.
citing papers explorer
-
Sparse Autoencoder Decomposition of Clinical Sequence Model Representations: Feature Complexity, Task Specialisation, and Mortality Prediction
Sparse autoencoders applied to a 14.5M-parameter clinical EHR model reveal progressive abstraction across layers, with SAE features outperforming dense ones for mortality in full-sequence probes but not in leakage-safe windows where dense representations match or exceed them.
-
FlatASCEND: Autoregressive Clinical Sequence Generation with Continuous Time Prediction and Association-Based Pharmacological Testing
FlatASCEND generates conditional clinical event sequences that partially recover known mechanistic drug associations from observational data but fail to maintain them under direct preference optimization and show weaker performance on longer outpatient timelines.