List your company

Medical Question Answering on MedQA (Accuracy, F1, Avg. Accuracy)

0.843AccuracyMediHive
0.709880.744440.7790.81356Mar 28, 2026

Evaluation results

MethodLinks
MediHiveAgent Configuration=Multi-Agent2026.030.8430.8250.814
MedAgentsAgent Configuration=Multi-Agent2026.030.8370.8210.803
Multiagent DebateAgent Configuration=Multi-Agent2026.030.8040.7830.793
Centralized Multi-Agent SystemAgent Configuration=Multi-Agent2026.030.7780.7610.766
Few-shot w/ CoTAgent Configuration=Single-Agent, Prompting Setting=Few-shot, Reasoning Strategy=CoT2026.030.7460.7310.735
Few-shot w/ CoT + SCAgent Configuration=Single-Agent, Prompting Setting=Few-shot, Reasoning Strategy=CoT + SC2026.030.7440.7290.741
Zero-shot w/ CoT + SCAgent Configuration=Single-Agent, Prompting Setting=Zero-shot, Reasoning Strategy=CoT + SC2026.030.7380.7090.73
Zero-shot w/ CoTAgent Configuration=Single-Agent, Prompting Setting=Zero-shot, Reasoning Strategy=CoT2026.030.7320.7050.72
Few-shotAgent Configuration=Single-Agent, Prompting Setting=Few-shot2026.030.7280.710.722
Zero-shotAgent Configuration=Single-Agent, Prompting Setting=Zero-shot2026.030.7150.6680.704