List your company

Question Answering on HEAD-QA (eval)

0Pathwise Violations (PathV)CSA
-4235077May 18, 2026

Evaluation results

MethodLinks
CSAN (EVAL size)=1,100, Err (Base verifier-fail rate)=26.0%, alpha* (Target risk level)=0.202026.05062.216.2
NEX-ConfN (EVAL size)=1,100, Err (Base verifier-fail rate)=26.0%, alpha* (Target risk level)=0.202026.050——
5 online baselinesN (EVAL size)=1,100, Err (Base verifier-fail rate)=26.0%, alpha* (Target risk level)=0.202026.05100——