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

CAPYBARA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Binary analysisCAPYBARA XRep
METEOR0.296
13
Binary analysisCAPYBARA Dec-Strip
METEOR Score18.2
13
Binary analysisCAPYBARA Dec-Anon
METEOR29.5
13
Binary analysisCAPYBARA (Dec-Full)
METEOR43.2
13
Binary AnalysisCAPYBARA XRep
EM16.63
13
Binary AnalysisCAPYBARA Dec-Strip
EM Score5.62
13
Binary AnalysisCAPYBARA Dec-Anon
Exact Match (EM)17.57
13
Binary AnalysisCAPYBARA Dec-Full
EM28.01
13
Binary AnalysisCAPYBARA XRep
ROUGE-L33.1
13
Binary AnalysisCAPYBARA Dec-Strip
ROUGE-L19.8
13
Binary AnalysisCAPYBARA Dec-Anon
ROUGE-L33.1
13
Binary AnalysisCAPYBARA Dec-Full
ROUGE-L47.7
13
LLM AlignmentCapybara
Win Rate28.34
8
Binary AnalysisCAPYBARA cross-representation
Performance (Full Mode)36.97
7
Showing 14 of 14 rows