Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Imputation on Semi-synthetic EHR dataset MNAR 80% (test)
Loading...
1
Mean Absolute Error (MAE)
GPT-5.4
0.8
2.15
3.5
4.85
May 6, 2026
Mean Absolute Error (MAE)
Root Mean Squared Error (RMSE)
Absolute Delta AC (|∆AC|)
Causal Structure Error (CSE)
Absolute Error on Imputed Y (|infY|)
MAE (Q1)
Error on Mean Outcome (Err_a_bar)
Absolute Delta ATE (|∆ATE|)
Mean Rank
Updated 27d ago
Evaluation Results
Method
Method
Links
Mean Absolute Error (MAE)
Root Mean Squared Error (RMSE)
Absolute Delta AC (|∆AC|)
Causal Structure Error (CSE)
Absolute Error on Imputed Y (|infY|)
MAE (Q1)
Error on Mean Outcome (Err_a_bar)
Absolute Delta ATE (|∆ATE|)
Mean Rank
GPT-5.4
framework=LLM-driven e...
2026.05
1
1
5
1
5
4
3
1
2.62
GPT-OSS-120b
framework=LLM-driven e...
2026.05
2
2
3
2
4
2
2
3
2.5
MissForest
2026.05
3
3
1
3
2
1
4
4
2.62
Qwen3.5-Plus
framework=LLM-driven e...
2026.05
4
4
4
4
3
5
5
5
4.25
CausalCFM
2026.05
5
5
2
6
1
6
6
6
4.62
LOCF
2026.05
6
6
6
5
6
3
1
2
4.38
Feedback
Search any
task
Search any
task