Out-Of-Domain
Benchmarks
TaskDatasetSOTA resultTrendResults
out-of-domain (test)
0.16Jensen-Shannon Divergence
56
Out-of-domain (OOD)
68.7mAP
28
Out-of-domain GSM-Hard, MultiArith, SVAMP
17GSM-Hard Accuracy
24
Out-of-domain heterogeneous public Internet recordings (test)
69.5Accuracy
24
Out-of-Domain (test)
81.8Agreement
15
Out-Of-Domain (OOD) (val)
11.08BLEU
10
Out-of-Domain
0.4Coherence
7
Out-of-Domain (OOD) Aggregate (eval)
8.4EER
6
Out-of-domain
0.007MSE
3
Out-of-domain
0.006MSE
3
Out-of-domain
0.212REL
3
Out-of-domain
0.087L1 Error
3