Sequence of tasks
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Sequence of 15 tasks (C-STANCE, FOMC, MeetingBank, Py150, ScienceQA, NumGLUE-cm, NumGLUE-ds, 20Minuten, dbpedia, amazon, yahoo, agnews, yelp, BoolQA, QQP) (long)
45.68OP (%)
10