ResearchBenchmarksDocstring-completion on Docstring-completion evaluation episodes (held-out)Follow2.195Policy ScoreMechRL2.085252.1401252.1952.249875May 25, 2026Evaluation ResultsMethodMethodLinksPolicy ScoreOracle ScoreGapMechRLK=1, In-episode picks=...K=1, In-episode picks=50, Evaluation episodes=≥ 202026.052.195—0.004OracleSelection=per-episode,...Selection=per-episode, Evaluation episodes=≥ 202026.05—2.198—