WikiHow
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
WikiHow
22.12ROUGE-2
26
WikiHow
18.61F1@5
19
WikiHow M4 benchmark (test)
100AUROC
12
WikiHow
38.4SummaC
12
WikiHow sampled (test)
17.58ROUGE
12
WikiHow (test)
39.06SummaC
12
WikiHow Golden (test)
91.03Acc
12
WikiHow (test)
19.23ROUGE
12
wikiHow (test)
55.4R@1
12
WikiHow
0.21ROUGE-1 Std Dev
8
WikiHow (test)
56.1ROUGE-1
7
WikiHow (test)
35.59ROUGE-1 Score
7
WikiHow
0.39Reward Score
6
WikiHow
4.45Conciseness
5
WikiHow 168k samples (test)
0.431ROUGE-1
5
WikiHow (test)
0.265ROUGE-1
4
WikiHow (test)
35.91ROUGE-1
3