Presentation by Susanne Kabatnik and Rukiye Smilla Burkart

XXII EURALEX INTERNATIONAL CONGRESS 2026: LEXICOGRAPHY IN THE AGE OF AI

The rapid development of artificial intelligence (AI) and natural language processing (NLP) has begun to transform the ways in which lexicographic data are compiled, analysed, and presented. Large language models, advanced corpus tools, and other machine learning methods offer unprecedented opportunities to accelerate tasks such as the extraction of lexical information, the induction of word senses, or the drafting of definitions. They also make possible new forms of user interaction, including conversational and personalized dictionaries that can adapt to different audiences or integrate multimodal content such as text, sound, and images. 

At the same time, these technologies pose serious challenges that go beyond purely technical concerns. Automatically generated content raises questions of reliability, bias, and transparency, while the integration of opaque systems into lexicographic workflows risks undermining scholarly standards of quality and accountability. The ecological impact of large-scale AI models must also be taken into consideration, raising doubts about their long-term sustainability in resource-intensive projects such as lexicography. Finally, the introduction of AI into the field invites reflection on the role of the lexicographer: how can human expertise remain central when automation is increasingly pervasive, and what kinds of collaborations between humans and machines are both effective and responsible? 

The conference Lexicography in the Age of AI will provide a forum for discussing these promises and risks. We will bring together professional lexicographers, publishers, researchers, software developers, and anyone interested in dictionaries to consider both the potential of AI for creating more dynamic, interactive dictionaries, and the ethical and ecological responsibilities that come with it. 

How Do You Define the (Known) Unknown? Prompt Engineering as a Method for AI-Assisted Meaning Definition in the Pandemictionary

2 Oct 2026, 10:40 | 01 | Sitzungssaal (01 | Sitzungssaal, OeAW Main Seat, 1st floor)

The formulation of definitions is a core task of lexicography, traditionally carried out manually. Large language models now offer new opportunities to automate this workflow. While AI-assisted definition generation has been well studied for well-known vocabulary, this is not yet the case for lesser-known historical vocabulary, which is scarcely represented in training data. This paper therefore investigates how far targeted prompt engineering can improve the quality of AI-generated definitions for historical pandemic vocabulary, using data from the cholera corpus developed within the Pandemictionary project. Eight headwords of varying frequency and semantic complexity are analysed across three prompt configurations, focusing on the role of corpus evidence in generating lexicographical definitions. The results show that AI systems can generate formally correct, lexicographically structured definitions. Prompts without evidence draw on general model knowledge and often produce anachronistic or hallucinated content, whereas evidence-based prompts yield definitions that are more historically accurate and firmly grounded in the corpus. Corpus evidence proves a crucial quality factor, especially for rare or poorly documented lexemes, though challenges remain regarding hallucinations, polysemy, and reducing definitions to their essential elements. The study highlights the potential of prompt engineering for digital lexicography while underscoring the continuing central role of human expertise in quality assurance.

____

For more information on the subproject „Pandemic vocabulary and LOD - (infectious) diseases in networked digital dictionaries,” please visit: https://www.uni-trier.de/forschung/forschungsfoerderung/landesforschungsfoerderung-1/loding/research-areas/tp1.