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

WikiHow

Benchmarks

Task NameDataset NameSOTA ResultTrend
Abstractive SummarizationWikiHow
ROUGE-222.12
26
Keyphrase ExtractionWikiHow
F1@518.61
19
Machine-generated text detectionWikiHow M4 benchmark (test)
AUROC100
12
Summarization FaithfulnessWikiHow
SummaC38.4
12
Abstractive SummarizationWikiHow sampled (test)
ROUGE17.58
12
Faithfulness EvaluationWikiHow (test)
SummaC39.06
12
Multimodal SequencingWikiHow Golden (test)
Acc91.03
12
SummarizationWikiHow (test)
ROUGE19.23
12
Step-goal linkingwikiHow (test)
R@155.4
12
SummarizationWikiHow
ROUGE-1 Std Dev0.21
8
Plan GenerationWikiHow (test)
ROUGE-156.1
7
Extractive SummarizationWikiHow (test)
ROUGE-1 Score35.59
7
SummarizationWikiHow
Reward Score0.39
6
SummarizationWikiHow
Conciseness4.45
5
Abstractive SummarizationWikiHow 168k samples (test)
ROUGE-10.4306
5
Unsupervised abstractive summarizationWikiHow (test)
ROUGE-10.265
4
Abstractive SummarizationWikiHow (test)
ROUGE-135.91
3
Showing 17 of 17 rows