Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

AISA-AR-FunctionCall

Benchmarks

Task NameDataset NameSOTA ResultTrend
Function CallingAISA-AR-FunctionCall Maghrebi 1.0 (test)
Function Name Accuracy61.58
2
Function CallingAISA-AR-FunctionCall Levantine 1.0 (test)
Function Name Accuracy69.48
2
Function CallingAISA-AR-FunctionCall Egyptian 1.0 (test)
Function Name Accuracy68.34
2
Function CallingAISA-AR-FunctionCall Gulf 1.0 (test)
Function Name Accuracy69.72
2
Function CallingAISA-AR-FunctionCall MSA 1.0 (test)
Function Name Accuracy76.13
2
Showing 5 of 5 rows