FAITHQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
FAITHQA (Misinterpretation)
29Perfect Score
14
FAITHQA (Misinterpretation)
22Perfect Score
7
FAITHQA (Omission)
60Perfect Score
7
FAITHQA (Omission)
38Perfect Score
7
FAITHQA (Omission)
57Perfect Score
7