List your company

ResearchFormerly WizWand

RA-QA

Benchmarks

TaskDatasetSOTA resultTrendResults

String-level response similarity

RA-QA Global, Discriminative tasks

0.9BERTScore

8

String-level response similarity

RA-QA Multiple-choice, Discriminative tasks

0.85BERTScore

4

String-level response similarity

RA-QA Single-Verify, Discriminative tasks

94BERTScore

4

Discriminative tasks

RA-QA

72Accuracy

4

Regression tasks

RA-QA

2.29MAE

3