Advanced search
2 files | 1.59 MB Add to list

Generative AI and lexicography : the current state of the art using ChatGPT

Author
Organization
Abstract
In this article, all ten papers and talks that have been devoted to the use of ChatGPT in lexicography so far are critically analysed, their results tabulated and cross-compared, from which the leading trends are determined. Extrapolating from the trendlines, a single short but robust new prompt is fine-tuned with which articles from different word classes are generated fully-automatically for a dictionary which compares favourably to the best practice in dictionary compilation. The conclusion is that a new age, that of the successful application of generative AI in lexicography, has dawned.
Keywords
Lexicography, Artificial Intelligence (AI), Large Language Model (LLM), Generative AI, Generative Pre-Trained Transformer (GPT), Chatbot, ChatGPT, Bard, Bing Chat, Claude, Stable Diffusion, Prompt, Bias, Hallucination, Black Box, Non-Deterministic Output, Memorisation, COBUILD, Full-sentence defining style, Authentic-like example sentences, BantUGent, LT3

Downloads

  • (...).pdf
    • supplementary material
    • |
    • UGent only
    • |
    • PDF
    • |
    • 575.08 KB
  • (...).pdf
    • full text (Published version)
    • |
    • UGent only
    • |
    • PDF
    • |
    • 1.01 MB

Citation

Please use this url to cite or link to this publication:

MLA
de Schryver, Gilles-Maurice. “Generative AI and Lexicography : The Current State of the Art Using ChatGPT.” INTERNATIONAL JOURNAL OF LEXICOGRAPHY, vol. 36, no. 4, 2023, pp. 355–87, doi:10.1093/ijl/ecad021.
APA
de Schryver, G.-M. (2023). Generative AI and lexicography : the current state of the art using ChatGPT. INTERNATIONAL JOURNAL OF LEXICOGRAPHY, 36(4), 355–387. https://doi.org/10.1093/ijl/ecad021
Chicago author-date
Schryver, Gilles-Maurice de. 2023. “Generative AI and Lexicography : The Current State of the Art Using ChatGPT.” INTERNATIONAL JOURNAL OF LEXICOGRAPHY 36 (4): 355–87. https://doi.org/10.1093/ijl/ecad021.
Chicago author-date (all authors)
de Schryver, Gilles-Maurice. 2023. “Generative AI and Lexicography : The Current State of the Art Using ChatGPT.” INTERNATIONAL JOURNAL OF LEXICOGRAPHY 36 (4): 355–387. doi:10.1093/ijl/ecad021.
Vancouver
1.
de Schryver G-M. Generative AI and lexicography : the current state of the art using ChatGPT. INTERNATIONAL JOURNAL OF LEXICOGRAPHY. 2023;36(4):355–87.
IEEE
[1]
G.-M. de Schryver, “Generative AI and lexicography : the current state of the art using ChatGPT,” INTERNATIONAL JOURNAL OF LEXICOGRAPHY, vol. 36, no. 4, pp. 355–387, 2023.
@article{01HDQA5VWRT4Q5KP1ZXD553BZG,
  abstract     = {{In this article, all ten papers and talks that have been devoted to the use of ChatGPT in lexicography so far are critically analysed, their results tabulated and cross-compared, from which the leading trends are determined. Extrapolating from the trendlines, a single short but robust new prompt is fine-tuned with which articles from different word classes are generated fully-automatically for a dictionary which compares favourably to the best practice in dictionary compilation. The conclusion is that a new age, that of the successful application of generative AI in lexicography, has dawned.}},
  author       = {{de Schryver, Gilles-Maurice}},
  issn         = {{0950-3846}},
  journal      = {{INTERNATIONAL JOURNAL OF LEXICOGRAPHY}},
  keywords     = {{Lexicography,Artificial Intelligence (AI),Large Language Model (LLM),Generative AI,Generative Pre-Trained Transformer (GPT),Chatbot,ChatGPT,Bard,Bing Chat,Claude,Stable Diffusion,Prompt,Bias,Hallucination,Black Box,Non-Deterministic Output,Memorisation,COBUILD,Full-sentence defining style,Authentic-like example sentences,BantUGent,LT3}},
  language     = {{eng}},
  number       = {{4}},
  pages        = {{355--387}},
  title        = {{Generative AI and lexicography : the current state of the art using ChatGPT}},
  url          = {{http://doi.org/10.1093/ijl/ecad021}},
  volume       = {{36}},
  year         = {{2023}},
}

Altmetric
View in Altmetric
Web of Science
Times cited: