REASONMAP
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
ReasonMap
55.1Accuracy
16
ReasonMap Plus
89.7Accuracy
16
REASONMAP-PLUS
88.95Weighted Accuracy
16
REASONMAP Long questions
62.5Weighted Accuracy
16
REASONMAP Short questions
0.6Weighted Accuracy
16
ReasonMap L (long questions)
0.075Weighted Accuracy
3
ReasonMap S (short questions)
15.44Weighted Accuracy
3