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

InfoBench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Instruction Following EvaluationInfoBench
Score86.1
23
Information FollowingInfoBench
Easy Score89.2
21
Reward ModelingInfoBench
Accuracy87.7
17
ClassificationInfoBench
Binary Accuracy92.8
12
ReasoningInfoBench
Easy Score89.2
11
Instruction FollowingInfoBench (test)
Score83.2
9
Instruction FollowingInfoBench
Accuracy85.2
8
Showing 7 of 7 rows