T0
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
T0 MTest11 P3 (test)
61.4Accuracy
42
T0 benchmark
85.8RTE
18
T0 Without SCloze dataset HyperT5 variant (test)
60.6Accuracy
14
T0 (test)
65.5Accuracy
8
T0 Taxonomy Evaluation Tasks (val)
59.1OBQA
7
T0 Evaluation Suite IA3 PEFT (held-out)
71.9RTE
6
T0 11B (test)
74.9Avg Test Score
6
T0
49Accuracy
5
T0 Zero-Shot
—Accuracy
0