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

2WikiMultiHop

Benchmarks

Task NameDataset NameSOTA ResultTrend
Continual routing2WikiMultiHop
Accuracy59.5
22
Question Answering2WikiMultiHop
EM38.4
11
Multi-hop Question Answering2WikiMultiHop (in-distribution)
Accuracy59.5
5
Showing 3 of 3 rows