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

Online A/B

Benchmarks

Task NameDataset NameSOTA ResultTrend
RecommendationOnline A/B (live production)
Emerging Creator Exposure Change (%)3
5
Access Control Flow ApprovalOnline A/B (test)
AUC4.9
2
RecommendationOnline A/B Item Scenario (test)
IPV3.64
1
Large-Scale Streaming RecommendationOnline A/B (test)
Pass-Through Rate45.04
1
Showing 4 of 4 rows