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

COGS

Benchmarks

Task NameDataset NameSOTA ResultTrend
Semantic ParsingCOGS (generalization)
Accuracy (Generalization)99
25
Semantic ParsingCOGS (test)
Exact Match Accuracy92.3
16
Semantic ParsingCOGS
Accuracy99.5
9
Compositional GeneralizationCOGS
Exact Match Accuracy83.9
6
Semantic ParsingCOGS nonce
Accuracy81.4
6
Showing 5 of 5 rows