ResearchBenchmarksMathematical Reasoning on RIMO-NFollow43.7Error RateSBBT43.58444.36745.1545.933May 26, 2026Evaluation ResultsMethodMethodLinksError RateObservation ScoreAUROC Gap ScoreBrier DeltaHC AUROC GapText/Self AUROC GapLT/Act AUROC GapSBBTBackbone=Qwen3-14B, Ev...Backbone=Qwen3-14B, Evaluation setting=Main2026.0543.718.5-0.065-0.1330.0740.045-0.011SBBTBackbone=14B, Evaluati...Backbone=14B, Evaluation setting=Main2026.0546.621.2-0.043-0.1630.0540.03-0.001