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

Real User Queries

Benchmarks

Task NameDataset NameSOTA ResultTrend
Retrieval-Augmented Generation1,000 Real User Queries 1.0 (full set)
CO4.404
6
Information RetrievalReal User Queries 1,000 (full)
CO4.404
3
Search EvaluationReal User Queries Complex (test)
NWR13
1
Search EvaluationReal User Queries Moderately Complex (test)
NWR5
1
Search EvaluationReal User Queries Simple (test)
NWR0
1
Showing 5 of 5 rows