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

Beaver

Benchmarks

Task NameDataset NameSOTA ResultTrend
Table RetrievalBEAVER
Capped Recall@2543.5
9
Table RetrievalBEAVER
Recall@1032.6
9
Table RetrievalBeaver (dev)
PR@517
8
Table RetrievalBEAVER
Recall@523.1
6
Safety ModerationBeaver Response
F1 Score84.4
5
Safety ModerationBeaver Prompt
F1 Score77.2
5
Text-to-SQLBEAVER
Execution Accuracy (EX)29.8
4
Showing 7 of 7 rows