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

C++

Benchmarks

Task NameDataset NameSOTA ResultTrend
Syntactic CorrectnessC++
Syntactic Correctness100
100
Student SimulationC++_5
Acc86
72
Grammar-Constrained DecodingC++
Relative Inference Time64.07
40
Functional CorrectnessC++
Success Rate U43.9
20
Structured GenerationC++
Mean Generation Latency (s)1.42
9
Adversarial Code ComplianceC++
Decoupling Probability100
9
Showing 6 of 6 rows