List your company

Expert-Iteration RLVR on MedQA, HEAD-QA, ARC-C, and CaseHOLD

4Pathwise Clean ScoreCSA
-0.160.9223.08May 18, 2026

Evaluation results

MethodLinks
CSAreplications per cell=20, alpha=pivotal alpha*2026.0544452
CRCreplications per cell=20, alpha=pivotal alpha*2026.054002.8
LTTreplications per cell=20, alpha=pivotal alpha*2026.0541114.2
ConfFactreplications per cell=20, alpha=pivotal alpha*2026.0531125.5
Fixed-Thresholdreplications per cell=20, alpha=pivotal alpha*2026.0524266.1
NEX-Confreplications per cell=20, alpha=pivotal alpha*2026.0514179.8
ACIreplications per cell=20, alpha=pivotal alpha*2026.0514190.7
SAOCPreplications per cell=20, alpha=pivotal alpha*2026.0504090.3
Naive-Tuningreplications per cell=20, alpha=pivotal alpha*2026.0504087.2
Always-Actreplications per cell=20, alpha=pivotal alpha*2026.05040100