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

VideoRewardBench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multimodal Reward ModelingVideoRewardBench
Macro Pairwise Accuracy68.2
18
Video Reward ModelingVideoRewardBench
Perception (long)76.2
16
Showing 2 of 2 rows