TASK
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Task
26.7BLEU
7
Task
84.4Accuracy
7
Task 7
0.76AUROC
6
Task 6
0.84AUROC
6
Task Setup 1
71.2SPL
5
Task 2
100OSR
4
Task Category D
4.44Functionally Diverse Responses
4
Task Category C
5Functionally Diverse Responses Count
4
Task 6
1.227TTFT (s)
4
Task 3
2.333TTFT (s)
4
Task 2
1.376TTFT (s)
4
Task #12
70Score
4
Task #11
0.87Score
4
Task #10
0.95Score
4
Task 9
0.65Score
4
Task #8
0.42Score
4
Task #6
84Score
4
Task #5
84Score
4
Task #4
97Score
4
Task #3
0.93Score
4
Task #2
27Score
4
Task #1
67Score
4
Task 8
72AUROC
4
TASK
0.692EM
4
Task 2 Boundary Conditions
52.18Minimum Compliance
3