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

Harmful Meme

Benchmarks

Task NameDataset NameSOTA ResultTrend
Safeguarding harmful image generationHarmful Meme Offensiveness
Average SSIM0.54
2
Safeguarding harmful image generationHarmful Meme Misogyny
Average SSIM0.9
2
Safeguarding harmful image generationHarmful Meme Racism
Average SSIM0.76
2
Safeguarding harmful image generationHarmful Meme Toxicity
SSIM0.24
1
Safeguarding harmful image generationHarmful Meme Sarcasm
Average SSIM0.82
1
Showing 5 of 5 rows