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

TofuEval-MeetingBank

Benchmarks

Task NameDataset NameSOTA ResultTrend
Factual Consistency DetectionTofuEval-MeetingBank Span level leakage-clean protocol (test)
AUC0.693
15
Factual Consistency DetectionTofuEval-MeetingBank Response level leakage-clean protocol (test)
AUC65.9
12
Showing 2 of 2 rows