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

MAUD

Benchmarks

Task NameDataset NameSOTA ResultTrend
Legal Contract RevisionMAUD
CQ Score86.89
25
Context PrioritizationMAUD
R@3375
12
Corpus poisoning detectionMAUD
Precision0.46
6
Evidence SelectionMAUD
Relative Evidence Count3.15
4
Evidence SelectionMAUD
F1 Score39
3
RetrievalMAUD
Recall@153.9
3
Sparse RetrievalMAUD
Recall@175.4
3
Showing 7 of 7 rows