Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Free-form answer generation on KLAR (OOD)

38.13Accuracy

GRPO

6.586814.775922.96531.1541Jun 4, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
38.13
2026.06
35.76
2026.06
30.75
2026.06
30.21
2026.06
29.96
2026.06
23.75
2026.06
17.6
2026.06
16.7
2026.06
13.3
2026.06
8.6
2026.06
8.3
2026.06
7.8