Literature
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Literature
72.7F1 Score
19
Literature-91
44.9Task Success Rate (TSR)
11
Literature factoid QA (test)
95.1Accuracy
9
Literature-91
41.4Top-1 Accuracy
3
Literature-91
74.6Top-1 Accuracy
3
Literature-91
93.6Top-1 Accuracy
3
Literature-91 Overall
42.9Pass@1
2
Literature-91 Extended 41 Complex Circuits
24.4Pass@1
2
Literature English
59.68F1 Score (%)
2