Eliciting Medical Reasoning with Knowledge-enhanced Data Synthesis: A Semi-Supervised Reinforcement Learning Approach
Open paper
Benchmark
Model
Rank
Results
multiple-choice-question-answering-mcqa-on-medmcqa
Qwen3-8B-Base + MedSSR
#6
Test Set (Acc-%): 0.6779
question-answering-on-pubmedqa
Llama-3.1-8B-Instruct + MedSSR
#29
Accuracy: 58.75
Rank counts only results with a code link.