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

K&K Logic Puzzles

Benchmarks

Task NameDataset NameSOTA ResultTrend
Logic ReasoningK&K Logic Puzzles OOD
Score Threshold 2 (OOD)99
25
Logic ReasoningK&K Logic Puzzles In-domain
Accuracy (Level 3)98
25
Logic PuzzlesK&K Logic Puzzles
Accuracy (Level 3)95
15
Logical ReasoningK&K Logic Puzzles
Accuracy (Level 3)96
12
Showing 4 of 4 rows