List your company

Medical Question Answering on PubMedQA (Accuracy, 95% CI, Avg. Accuracy)

77.6AccuracyPeer-reviewd reasoning
56.17661.73867.372.862Jun 13, 2026

Evaluation results

MethodLinks
Peer-reviewd reasoningModels=(1) + (3) + (4) + (5)2026.0677.673.882
Chain-of-thoughtModel=Llama-3.1-8B2026.0675.471.671.1
Chain-of-thoughtModel=Qwen2.5-7B2026.066762.865
Chain-of-thoughtModel=Deepseek-llm-7b2026.065752.643.7