LLM-Driven Evolutionary Imputation
Cross-source consensus on LLM-Driven Evolutionary Imputation from 1 sources and 7 claims.
1 sources · 7 claims
Benefits
Comparisons
Highlighted claims
- The imputation baselines included LLM-driven evolutionary imputers using GPT-5.4, Qwen3.5-Plus, and GPT-OSS-120b. — Joint Treatment Effect Estimation from Incomplete Healthcare Data: Temporal Causal Normalizing Flows with LLM-driven Evolutionary MNAR Imputation
- GPT-5.4 achieved the best pooled mean rank among imputation methods. — Joint Treatment Effect Estimation from Incomplete Healthcare Data: Temporal Causal Normalizing Flows with LLM-driven Evolutionary MNAR Imputation
- GPT-5.4's pointwise reconstruction degraded only modestly as missingness increased from 30% to 80%. — Joint Treatment Effect Estimation from Incomplete Healthcare Data: Temporal Causal Normalizing Flows with LLM-driven Evolutionary MNAR Imputation
- GPT-5.4 had the best pointwise MAE and RMSE at 30% missingness. — Joint Treatment Effect Estimation from Incomplete Healthcare Data: Temporal Causal Normalizing Flows with LLM-driven Evolutionary MNAR Imputation
- GPT-5.4 had the best pointwise MAE and RMSE at 50% missingness. — Joint Treatment Effect Estimation from Incomplete Healthcare Data: Temporal Causal Normalizing Flows with LLM-driven Evolutionary MNAR Imputation
- GPT-5.4 had the best pointwise MAE and RMSE at 80% missingness and an ATE residual of 0.013. — Joint Treatment Effect Estimation from Incomplete Healthcare Data: Temporal Causal Normalizing Flows with LLM-driven Evolutionary MNAR Imputation
- The evolutionary loop materially improved GPT-5.4 imputation performance, especially at 50% to 80% missingness. — Joint Treatment Effect Estimation from Incomplete Healthcare Data: Temporal Causal Normalizing Flows with LLM-driven Evolutionary MNAR Imputation