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

QuaRTz

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multi-hop Open-domain Question AnsweringQuartz
Accuracy94.1
6
Question AnsweringQuartz (test)
Accuracy90.2
4
Ranking correlation with full dataset evaluationQuaRTz
Kendall Correlation0.37
4
Showing 3 of 3 rows