ResearchBenchmarksMathematical Reasoning on AIME '25 (Accuracy, Speedup)Follow26.67AccuracyARROL19.733221.534123.33525.1359Mar 25, 2026Evaluation ResultsMethodMethodLinksAccuracySpeedupARROLModel=Qwen-3-1.7B-Base...Model=Qwen-3-1.7B-Base, Base Method=DAPO2026.0326.671.7DAPOModel=Qwen-3-1.7B-BaseModel=Qwen-3-1.7B-Base2026.0320—