Joint Treatment Effect Estimation from Incomplete Healthcare Data: Temporal Causal Normalizing Flows with LLM-driven Evolutionary MNAR Imputation

<|message_model|><|content_text|>A two-stage pipeline combines exact invertible counterfactual inference via DAG-constrained temporal normalizing flows with LLM-driven evolutionary MNAR imputation to estimate treatment effects from incomplete longitudinal EHRs. On synthetic benchmarks and Swiss diab

Olivia Jullian Parra, Sara Zoccheddu, David C Cerezo, Tom Forzy and 6 more

Paris Poster Session 6, Fri, Dec 11, 2:30 PM–4:30 PM, Paris Poster Hall · Published 2026