ResearchBenchmarksCode on MBPP (Avg@4)Follow90.24Avg@4AIPO80.235282.832685.4388.0274May 8, 2026Evaluation ResultsMethodMethodLinksAvg@4AIPOModel=Qwen2.5-32B-Inst...Model=Qwen2.5-32B-Instruct2026.0590.24GRPOModel=Qwen2.5-32B-Inst...Model=Qwen2.5-32B-Instruct2026.0586.71SFTModel=Qwen2.5-32B-Inst...Model=Qwen2.5-32B-Instruct2026.0583.57OriginalModel=Qwen2.5-32B-Inst...Model=Qwen2.5-32B-Instruct2026.0580.62