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

GenAI

Benchmarks

Task NameDataset NameSOTA ResultTrend
Global Multimodal Question AnsweringGenAI
Comprehensiveness95.2
12
Video Reward ModelingGenAI
Accuracy (w/ Tie)54.8
7
Aesthetic Preference PredictionGenAI
Accuracy (w/ Ties)46.29
6
Region SegmentationGenAI
ARI50.92
4
Intra-image patch-level matchingGenAI
AP88.42
2
Region MatchingGenAI
Region Accuracy78.43
2
Cross-image patch-level matchingGenAI
AP83.49
1
Attack-Mitigation EffectivenessGenAI
CIS1
1
Showing 8 of 8 rows