Federated generative event models trained across three independent ICU datasets match most centralized multi-site performance and are substantially more transportable across hospitals than LightGBM.
An empirical analysis of compute-optimal large language model training,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Federated generative event models for tokenized electronic health records
Federated generative event models trained across three independent ICU datasets match most centralized multi-site performance and are substantially more transportable across hospitals than LightGBM.