Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Natural Language Generation on XSUM 40K
Loading...
13
BLEU
BART-L
10.608
11.229
11.85
12.471
May 3, 2023
BLEU
ROUGE
BScore
PPL
Updated 3mo ago
Evaluation Results
Method
Method
Links
BLEU
ROUGE
BScore
PPL
BART-L
FLOPS=19.6, Latency=25...
2023.05
13
31.1
73.9
1.7
BART-KD
FLOPS=5.1, Latency=68,...
2023.05
12.3
30.2
73.5
1.9
T5-L
FLOPS=38.7, Latency=53...
2023.05
11.5
29.3
72.7
1.7
T5-KD
FLOPS=2.7, Latency=144...
2023.05
10.7
28.2
71.8
1.9
Feedback
Search any
task
Search any
task
