Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Summarization

Benchmarks

Task NameDataset NameSOTA ResultTrend
SummarizationSummarization
ROUGE-L26.77
18
SummarizationSummarization dataset
ROUGE-L F167.8
16
SummarizationSummarization
Edit Distance6,573
12
SummarizationSummarization (Weak User)
Mean Total Edit Distance0.2577
10
SummarizationSummarization (Strong User)
Mean Total Edit Distance0.1845
10
SummarizationSummarization
Rouge-L18.4
10
SummarizationSummarization
Grade73.18
6
SummarizationSummarization Human Evaluation (test)
Consistency4
6
Critique Quality EvaluationSummarization
Win Rate75
6
FaithfulnessSummarization (test)
Reward-0.268
4
Label aggregation assessmentSummarization (test)
Test Accuracy68
4
SummarizationSummarization (test)
Rank 1 Frequency0.52
4
Showing 12 of 12 rows