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

S-NI

Benchmarks

Task NameDataset NameSOTA ResultTrend
Instruction FollowingS-NI
Rouge-L38.7
119
Instruction FollowingS-NI
SBERT Similarity60.2
24
Instruction FollowingS-NI (OOD)
Rouge-L36.1
20
Dialogue GenerationS-NI
Rouge-L19.82
16
Instruction FollowingS-NI
Instruction Following Score63.8
5
Showing 5 of 5 rows