Languages Through the Looking Glass of BPE Compression
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F23%3ANIA5XQEB" target="_blank" >RIV/00216208:11320/23:NIA5XQEB - isvavai.cz</a>
Result on the web
<a href="https://direct.mit.edu/coli/article/doi/10.1162/coli_a_00489/116635" target="_blank" >https://direct.mit.edu/coli/article/doi/10.1162/coli_a_00489/116635</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1162/coli_a_00489" target="_blank" >10.1162/coli_a_00489</a>
Alternative languages
Result language
angličtina
Original language name
Languages Through the Looking Glass of BPE Compression
Original language description
"Byte-pair encoding (BPE) is widely used in NLP for performing subword tokenization. It uncovers redundantpatterns for compressing the data, and hence alleviates the sparsity problem in downstream applications.Subwords discovered during the first merge operations tend to have the most substantial impact on thecompression of texts."
Czech name
—
Czech description
—
Classification
Type
J<sub>ost</sub> - Miscellaneous article in a specialist periodical
CEP classification
—
OECD FORD branch
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
Result continuities
Project
—
Continuities
—
Others
Publication year
2023
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Name of the periodical
"Computational Linguistics"
ISSN
1530-9312
e-ISSN
—
Volume of the periodical
""
Issue of the periodical within the volume
2023
Country of publishing house
US - UNITED STATES
Number of pages
59
Pages from-to
1-59
UT code for WoS article
—
EID of the result in the Scopus database
—