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

Flan

Benchmarks

Task NameDataset NameSOTA ResultTrend
Instruction TuningFLAN subset Average (test)
ROUGE-156.7
7
Instruction TuningFLAN All (test)
ROUGE-157.5
7
Instruction TuningFLAN Dual (test)
ROUGE-156.3
7
Instruction TuningFLAN Single (test)
ROUGE-156.2
7
Natural Language ProcessingFLAN 8-task subset: arc_challenge, cosmos_qa, definite_pronoun_resolution, glue_qqp, hellaswag, mnli, squad_v1, sst2
Closed-book QA71
7
Instruction FollowingFlan
Paraphrase0.7549
2
Showing 6 of 6 rows