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

Chain-of-thought reasoning on EUREQA (held-out half of hard_5)

21.6Best@3

Random-w

7.5611.20514.8518.495May 21, 2026
Updated 12d ago

Evaluation Results

MethodLinks
2026.05
21.622.222.923.710.5
2026.05
21.323.625.727.951.2
2026.05
21.221.922.623.611.9
2026.05
212324.926.752.6
2026.05
20.921.622.423.711.7
2026.05
20.621.422.423.714
2026.05
10.211.312.61417.1
2026.05
8.19.611.715.317.1