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

First-Order Logic Reasoning on LogicNLI

76.6Pass@1

SFT w/ Long CoT

47.89655.34862.870.252Jun 10, 2025
Updated 4d ago

Evaluation Results

MethodLinks
2025.06
76.6-
2025.06
75.4-
2025.06
74.8-
2025.06
73.8-
2025.06
70.40.1
2025.06
66.6-
2025.06
66.6-
2025.06
66.60.4
2025.06
65.8-
2025.06
64.4-
2025.06
64.4-
2025.06
60-
2025.06
58-
2025.06
55-
2025.06
55-
2025.06
54-
2025.06
54-
2025.06
49-