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

HUB

Benchmarks

Task NameDataset NameSOTA ResultTrend
Sarcasm RankingHUB
Ranking Score65.5
9
Sarcasm MatchingHUB
Matching Rate64.5
9
Concept UnlearningHUB NSFW category
Target51.5
7
Concept UnlearningHUB Celebrity category
Target Score68.5
7
Concept UnlearningHUB IP category
Target Success85.5
7
Concept UnlearningHUB Style category
Target Metric65.5
7
Hallucination DetectionHUB (test)
Algorithmic76.8
7
Showing 7 of 7 rows