IFBench
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
IFBench
27Pass@1
6
IFBench
9.7Genuine Followup Rate
6
IFBench
77.1Score
6
IFBench
33.3Utility Score
5
IFBench Out-of-Distribution
41.5Initial F1
5
IFBench In-Distribution
41.1Initial F1
5
IFBench
72.2Loose Prompt Success Rate
4
IFBench
2,438Max System Prompt Token Length
4
IFBench
46.33Score
4
IFBench (val)
90.6Score
4
IFBench
73.8Prompt Loose Score
4
IFBench Strict
31.5Avg@10
2
IFBench
71.9Score
1