Various
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Various (test)
3.27PESQ
11
Various
89.1F1 Score
1
Various SCM, HotpotQA, MS-MARCO, SQuAD, LongBench, Ruler, InfiniteBench, LOCOMO, LOCCO
21.45Execution Time Reduction
1
Various Comparison of published systems
—Primary metric
0