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

CCBench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Cultural Multimodal UnderstandingCCBench
Score27.9
20
Chinese Culture Multimodal EvaluationCCBench (dev)
Accuracy71.2
12
Multimodal UnderstandingCCBench 80% (test)
Accuracy73.96
10
Multimodal Question AnsweringCCBench
Score41.2
9
Multimodal Chinese Cultural Knowledge UnderstandingCCBench (test)
Average Score47.6
9
Showing 5 of 5 rows