List your company

Evaluation of metrics' alignment with human judgment on MedVal-Bench

0.933Spearman CorrelationAgentsEval
0.094760.312380.530.74762Jan 23, 2026

Evaluation results

MethodLinks
AgentsEval2026.010.9330.846
Agent(det)2026.010.9321.345
Agent(sim)2026.010.9310.934
B-Score2026.010.7821.18
ROUGE-12026.010.6054.43
METEOR2026.010.4115.099
BLEU2026.010.3397.636
CHRF2026.010.1274.264