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

HealthVer

Benchmarks

Task NameDataset NameSOTA ResultTrend
Fact-checkingHealthVer
F1-macro68
21
Natural Language Explanation GenerationHealthVer (test)
Faithfulness0.033
9
Human Evaluation of ExplanationsHealthVer
Helpfulness (MAR)2.15
5
Knowledge Poisoning AttackHealthVer k=10 (test)
ASR21
4
Claim VerificationHealthVer (test)
Precision (NE)47
2
Showing 5 of 5 rows