Summarization
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Summarization
95Win Rate
39
Summarization
32.1PRR
32
Summarization
44.99Rouge-L
23
Summarization
49ROUGE-1 Score
21
Summarization dataset
67.8ROUGE-L F1
16
Summarization
6,573Edit Distance
12
Summarization (Weak User)
0.258Mean Total Edit Distance
10
Summarization (Strong User)
0.185Mean Total Edit Distance
10
Summarization
89Preference Labeling Accuracy
8
Summarization
73.18Grade
6
Summarization Human Evaluation (test)
4Consistency
6
Summarization
75Win Rate
6
Summarization (test)
-0.268Reward
4
Summarization (test)
68Test Accuracy
4
Summarization (test)
0.52Rank 1 Frequency
4
Summarization Low Tier
13.9R Score
3
Summarization High Tier
51.8R Score
3
Summarization
36ROUGE-1
3