Transformer-XL: Attentive Language Models beyond a Fixed-Length Context
39675Knowledge Triples
| Source Entity | Relationship | Target Entity |
|---|---|---|
|
Transformer XL
Method |
evaluatedIn |
enwiki8
Material |
|
Transformer XL
Method |
evaluatedIn |
One Billion Word
Material |
|
Transformer XL
Method |
evaluatedIn |
text8
Material |
|
Transformer XL
Method |
evaluatedIn |
WikiText 103
Material |
|
Transformer XL
Method |
improves |
bpc/perplexity
Metric |