Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context
URI: https://gptkb.org/entity/Transformer-XL:_Attentive_Language_Models_Beyond_a_Fixed-Length_Context
GPTKB entity
Statements (29)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
gptkb:academic_journal
|
| gptkbp:arXivID |
1901.02860
|
| gptkbp:author |
gptkb:Jaime_Carbonell
gptkb:Ruslan_Salakhutdinov gptkb:Yiming_Yang gptkb:Zhilin_Yang gptkb:Zihang_Dai gptkb:Quoc_V._Le |
| gptkbp:bench |
gptkb:WikiText-103
One Billion Word enwik8 |
| gptkbp:citation |
high (over 2000 citations as of 2024)
|
| gptkbp:contribution |
improves language modeling performance
introduces novel positional encoding scheme introduces segment-level recurrence mechanism enables learning dependency beyond fixed-length context |
| gptkbp:field |
gptkb:machine_learning
deep learning natural language processing |
| gptkbp:improves |
previous state-of-the-art models on language modeling benchmarks
|
| gptkbp:influenced |
subsequent transformer architectures
|
| gptkbp:language |
English
|
| gptkbp:proposedBy |
Transformer-XL model
|
| gptkbp:publicationYear |
2019
|
| gptkbp:publishedIn |
gptkb:arXiv
|
| gptkbp:repository |
https://github.com/kimiyoung/transformer-xl
|
| gptkbp:bfsParent |
gptkb:Transformer-XL
|
| gptkbp:bfsLayer |
8
|
| https://www.w3.org/2000/01/rdf-schema#label |
Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context
|