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

Contract NLI

Benchmarks

Task NameDataset NameSOTA ResultTrend
Context PrioritizationContract-NLI
Recall@3100
12
Natural Language InferenceContract-NLI (test)
macro-F10.86
12
Corpus poisoning detectionContract-NLI
Precision (Detection)69
6
Evidence SelectionContract-NLI
Relative Evidence Count21.33
4
Evidence SelectionContract-NLI
F1 Score51
3
Natural Language InferenceContract NLI SCROLLS (val)
Exact Match77.7
2
Showing 6 of 6 rows