Electronic lexicography in the 21st century (eLex 2025) : Intelligent lexicography : Proceedings of the eLex 2025 conference

Abstract

Conference topic: Intelligent Lexicography - With the emergence of Large Language Models (LLMs) and a growing array of Artificial Intelligence (AI) tools, lexicographers now have unprecedented opportunities to enhance the efficiency and depth of language analysis and description. These tools offer new approaches to language analysis and description, potentially transforming traditional lexicographic workflows. However, harnessing the full potential of these technologies requires an approach we term intelligent lexicography. This approach calls for a systematic examination of both the benefits and limitations of LLMs in lexicographic applications. While LLMs can excel at processing vast amounts of linguistic data, challenges such as data quality, ethical issues, non-replicability, and the risk of over-reliance on automated processes must be addressed. Moreover, intelligent lexicography seeks to go beyond merely using AI tools; it aims to refine and adapt these technologies to better meet the specific needs of lexicographers. This includes developing methods to ensure that AI-generated outputs align with lexicographic principles and exploring ways to make AI even more effective and contextually aware from a lexicographic perspective. Finally, intelligent lexicography should also investigate ways in which lexicographic data can be used in the development of LLMs.