List your company

Question Answering on MedQA (Q, Ealign)

83.89Q ScoreGRPO
58.316464.955771.59578.2343Jun 12, 2026

Evaluation results

MethodLinks
GRPOBackbone Model=Ministral 3 8B, Evaluator=GPT-OSS 20B2026.0683.8916.28
CoRABackbone Model=Ministral 3 8B, Evaluator=GPT-OSS 20B2026.0682.1716.87
SFTBackbone Model=Ministral 3 8B, Evaluator=GPT-OSS 20B2026.0681.9217.48
BaseBackbone Model=Ministral 3 8B, Evaluator=GPT-OSS 20B2026.0681.3118.15
CoRABackbone Model=Gemma 2 9B, Evaluator=GPT-OSS 20B2026.0678.7220.42
GRPOBackbone Model=Gemma 2 9B, Evaluator=GPT-OSS 20B2026.0672.0126.84
SFTBackbone Model=Gemma 2 9B, Evaluator=GPT-OSS 20B2026.0671.5428.48
BaseBackbone Model=Gemma 2 9B, Evaluator=GPT-OSS 20B2026.0669.8329.28
GRPOBackbone Model=Qwen2.5 7B, Evaluator=GPT-OSS 20B2026.0664.8928.56
CoRABackbone Model=Qwen2.5 7B, Evaluator=GPT-OSS 20B2026.0663.7832.29
SFTBackbone Model=Qwen2.5 7B, Evaluator=GPT-OSS 20B2026.0661.2634.79
BaseBackbone Model=Qwen2.5 7B, Evaluator=GPT-OSS 20B2026.0659.336.01