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

PrivacyQA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Legal Contract RevisionPrivacyQA
CQ87.75
25
Legal TaskPrivacyQA
Performance67.8
18
Context PrioritizationPrivacyQA
Recall@6100
12
Answer Sentence SelectionPrivacyQA 1.0 (test)
SAE0.611
12
Corpus poisoning detectionPrivacyQA
Precision60
6
Evidence SelectionPrivacyQA
Relative Evidence Count3.33
4
Evidence SelectionPrivacyQA
F148
3
Showing 7 of 7 rows