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

Summarization on LongBench GovReport (ROUGE-L)

34.61ROUGE-L

Baseline

16.451621.165825.8830.5942Feb 13, 2026Feb 15, 2026Feb 17, 2026Feb 19, 2026Feb 21, 2026Feb 23, 2026Feb 26, 2026
Updated 11d ago

Evaluation Results

MethodLinks
2026.02
34.61
2026.02
34.47
2026.02
34.33
2026.02
34.33
2026.02
33.86
2026.02
33.8
2026.02
33.76
2026.02
33.62
2026.02
33.38
2026.02
32.73
2026.02
32.25
2026.02
32.11
2026.02
29.27
2026.02
28.89
2026.02
28.87
2026.02
28.57
2026.02
27.82
2026.02
27.29
2026.02
27.1
2026.02
27.01
2026.02
26.88
2026.02
26.72
2026.02
26.7
2026.02
26.68
2026.02
26.64
2026.02
26.38
2026.02
26.12
2026.02
25.69
2026.02
25.18
2026.02
21
2026.02
20.6
2026.02
20.5
2026.02
20.4
2026.02
20.3
2026.02
19.8
2026.02
19.7
2026.02
19.1
2026.02
17.15