Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Medical Multi-choice Question Answering on MMedBench (test)

0.1494Token Perplexity (log)

OFA-KD

0.148820.1527350.156650.160565Feb 15, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.02
0.1494
2026.02
0.1494
2026.02
0.151
2026.02
0.1526
2026.02
0.1535
2026.02
0.1546
2026.02
0.1551
2026.02
0.1557
2026.02
0.1569
2026.02
0.1571
2026.02
0.1573
2026.02
0.1574
2026.02
0.1588
2026.02
0.1603
2026.02
0.1609
2026.02
0.1639