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

AlpacaFact

Benchmarks

Task NameDataset NameSOTA ResultTrend
Long-form Question AnsweringAlpacaFact FullP (100% polluted)
VeriScore F1@k78.65
26
Long-form Question AnsweringAlpacaFact MixP (50% polluted)
VeriScore F1@k81.7
26
Long-form Question AnsweringAlpacaFact
VeriScore F166.9
14
Showing 3 of 3 rows