List your company

LLM win-rate estimation ranking on LLM benchmark (Appendix)

1Spearman CorrelationAIPW-EM
-0.040.230.50.77Apr 23, 2026

Evaluation results

MethodLinks
AIPW-EMn=252026.0411000
AutoCaln=252026.0411000
Labeled-onlyn=252026.0411000
LinearCaln=252026.0411000
PPI++n=252026.0411000
Labeled-onlyn=502026.0411000
PPI++n=502026.040.7587.50.003
AIPW-EMn=502026.040.5750.006
AutoCaln=502026.040.5750.006
LinearCaln=502026.040.5750.006
AIPWn=252026.040.2562.50.01
PPIn=252026.040.2562.50.01
AIPWn=502026.040.2562.50.01
PPIn=502026.040500.013