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

Abstract

Benchmarks

Task NameDataset NameSOTA ResultTrend
Image Captioning EvaluationAbstract-50S
Mean Accuracy76.9
4
Retrieval-Augmented GenerationAbstract single
F1 Score29
3
Concept Erasure AttackAbstract
LPIPS0.44
3
Showing 3 of 3 rows