https://arxiv.org/abs/2304.12404

Skip to main content
Cornell University
We gratefully acknowledge support from
the Simons Foundation and member institutions.
 
arxiv logo > cs > arXiv:2304.12404
[                    ]

Help | Advanced Search

[All fields        ]
Search
arXiv logo
Cornell University Logo
[                    ] GO
quick links

  * Login
  * Help Pages
  * About

Computer Science > Computation and Language

arXiv:2304.12404 (cs)
[Submitted on 24 Apr 2023]

Title:Semantic Tokenizer for Enhanced Natural Language Processing

Authors:Sandeep Mehta, Darpan Shah, Ravindra Kulkarni, Cornelia
Caragea
Download a PDF of the paper titled Semantic Tokenizer for Enhanced
Natural Language Processing, by Sandeep Mehta and 3 other authors
Download PDF

    Abstract: Traditionally, NLP performance improvement has been
    focused on improving models and increasing the number of model
    parameters. NLP vocabulary construction has remained focused on
    maximizing the number of words represented through subword
    regularization. We present a novel tokenizer that uses semantics
    to drive vocabulary construction. The tokenizer includes a
    trainer that uses stemming to enhance subword formation. Further
    optimizations and adaptations are implemented to minimize the
    number of words that cannot be encoded. The encoder is updated to
    integrate with the trainer. The tokenizer is implemented as a
    drop-in replacement for the SentencePiece tokenizer. The new
    tokenizer more than doubles the number of wordforms represented
    in the vocabulary. The enhanced vocabulary significantly improves
    NLP model convergence, and improves quality of word and sentence
    embeddings. Our experimental results show top performance on two
    Glue tasks using BERT-base, improving on models more than 50X in
    size.

Subjects: Computation and Language (cs.CL)
Cite as:  arXiv:2304.12404 [cs.CL]
          (or arXiv:2304.12404v1 [cs.CL] for this version)
          https://doi.org/10.48550/arXiv.2304.12404
          Focus to learn more
          arXiv-issued DOI via DataCite

Submission history

From: Sandeep Mehta [view email]
[v1] Mon, 24 Apr 2023 19:33:41 UTC (405 KB)
Full-text links:

Download:

    Download a PDF of the paper titled Semantic Tokenizer for
    Enhanced Natural Language Processing, by Sandeep Mehta and 3
    other authors
  * PDF
  * Other formats

[by-nc-sa-4]
Current browse context:
cs.CL
< prev   |   next >
new | recent | 2304
Change to browse by:
cs

References & Citations

  * NASA ADS
  * Google Scholar
  * Semantic Scholar

a export BibTeX citation Loading...

BibTeX formatted citation

x
[loading...          ]
Data provided by:

Bookmark

BibSonomy logo Mendeley logo Reddit logo ScienceWISE logo
(*) Bibliographic Tools

Bibliographic and Citation Tools

[ ] Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
[ ] Litmaps Toggle
Litmaps (What is Litmaps?)
[ ] scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
( ) Code, Data, Media

Code, Data and Media Associated with this Article

[ ] DagsHub Toggle
DagsHub (What is DagsHub?)
[ ] Links to Code Toggle
Papers with Code (What is Papers with Code?)
[ ] ScienceCast Toggle
ScienceCast (What is ScienceCast?)
( ) Demos

Demos

[ ] Replicate Toggle
Replicate (What is Replicate?)
[ ] Spaces Toggle
Hugging Face Spaces (What is Spaces?)
( ) Related Papers

Recommenders and Search Tools

[ ] Link to Influence Flower
Influence Flower (What are Influence Flowers?)
[ ] Connected Papers Toggle
Connected Papers (What is Connected Papers?)
[ ] Core recommender toggle
CORE Recommender (What is CORE?)

  * Author
  * Venue
  * Institution
  * Topic

( ) About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and
share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have
embraced and accepted our values of openness, community, excellence,
and user data privacy. arXiv is committed to these values and only
works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community?
Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is
MathJax?)

  * About
  * Help

  * Click here to contact arXiv Contact
  * Click here to subscribe Subscribe

  * Copyright
  * Privacy Policy

  * Web Accessibility Assistance
  * arXiv Operational Status
    Get status notifications via email or slack