LongForm
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
longform_qa
100Accuracy
48
LongForm QA
99.88Accuracy
24
Longform
0.221PRR
18
Longform
65.9AUROC
18
LongForm Bio
0.857FactScore
12
Longform
0.654AUROC
9
Longform
21.9Generation Metric
6
LongForm
33.4Precision
6
LongForm
100Score (gpt-5.1)
5
LongForm
100Score (gpt-5.1)
5