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

Crosswords

Benchmarks

Task NameDataset NameSOTA ResultTrend
Explorative ReasoningCrosswords Word-level (test)
Accuracy13.5
11
Explorative ReasoningCrosswords Letter-level (test)
Accuracy23.4
11
Word Puzzle SolvingCrosswords Word
Accuracy13.5
11
Word Puzzle SolvingCrosswords Letter
Accuracy23.4
11
Mini CrosswordsMini Crosswords GooBix (test)
Letter Success Rate82.4
6
ReasoningCrosswords
Inference Time (seconds)37.09
4
Showing 6 of 6 rows