The 1st International Conference on Lexicology and Lexicography (LexiKonf 2025), held in Budapest from 29 September to 1 October 2025, marked an important milestone for the lexicographic community in Hungary and abroad. Organised by the Institute for Lexicology of the Eötvös Loránd University (ELTE) Research Centre for Linguistics, this second iteration of the LexiKonf series fulfilled a long-standing ambition: to transform a formerly national event into a genuinely international forum. Thanks to the support provided by a national public funding scheme,1 the conference succeeded in bringing together participants from several continents, offering a diverse, intergenerational, and methodologically varied overview of current research in lexicology and lexicography.
This development is particularly significant in the Hungarian context. Prior to the launch of LexiKonf in 2023, no conference series dedicated specifically to lexicography or lexicology had existed in Hungary for decades—despite the country’s strong institutional traditions in dictionary-making, including long-term projects such as the Comprehensive Dictionary of Hungarian, the New Etymological Dictionary of Hungarian, and the New Dialect Dictionary of Hungarian. The intention behind establishing LexiKonf as a recurring event is to provide a regular meeting place for researchers, strengthen professional networks, and support the renewal and visibility of lexicographic work.
The plenary lectures reflected this balance between continuity and innovation. On the opening day, Sven Tarp (Aarhus University) discussed how recent developments in generative AI are reshaping the landscape of lexicographic products. In his talk »Do Lexicography a Favour: Stop Focusing on Dictionaries«, he argued that the traditional dictionary is losing its central role as users increasingly rely on problem-oriented, digitally mediated forms of linguistic assistance. Building on the distinction between dictionography and glossography, he presented AI-generated glosses developed for an L2 writing assistant, where errors in learner texts trigger brief suggestions, short explanations, or extended grammatical notes. These dynamic, non-lemma-based glosses illustrate how lexicographic support can become more contextualised and personalised, pointing towards interactive, AI-supported glossography as a potential direction for future lexicographic practice.
The second plenary, delivered by Iztok Kosem (University of Ljubljana), presented ongoing lexicographic work at the Centre for Language Resources and Technologies, with a particular focus on developing the Digital Dictionary Database of Slovene. In his talk, »Common Sense(s) in Slovene Lexicography«, he outlined the challenges involved in merging heterogeneous resources, linking external materials such as crowdsourced data, and defining shared criteria for representing senses and concepts. Kosem also discussed practical issues arising from creating resources from scratch, the transition from the Lexonomy platform to an in-house database editor and its implications for editorial workflows, as well as initial experiments with integrating large language models into the lexicographic process. He concluded with examples showing how the emerging database is already being applied beyond lexicography.
The final morning featured a plenary lecture by Ana Salgado (Lisbon Academy of Sciences), who reviewed a decade of developments in the Dicionário da Língua Portuguesa project. Her talk described the transition from traditional editorial practice to corpus-based methods, the redefinition of microstructural components, and the creation of a sustainable digital platform. She also outlined recent steps towards integrating new tools and procedures into the lexicographic workflow, including approaches to updating content in response to linguistic change, the use of user log data, semi-automatic methods for identifying neologisms, and considerations related to language policy. By situating these developments within institutional and technological constraints, the presentation invited reflection on how academy dictionaries can remain scientifically rigorous and socially relevant.
With two parallel sessions and 55 presentations, LexiKonf 2025 showcased the breadth of contemporary lexicographic inquiry. The sessions revealed several shared concerns across languages, institutions, and methodological backgrounds. One group of presentations focused on construction-based approaches to linguistic modelling and their lexicographic implications. Bálint Sass (ELTE Research Centre for Linguistics) introduced the use of morphological dependency trees in the developing Hungarian Constructicon, illustrating how constructions can be represented through hierarchical morphological relations. Edmond Cane (Beijing International Studies University) presented a constructionist framework for corpus processing and dictionary modelling, in which all levels of linguistic structure—from basic lemmas to larger phrasal units—are treated as constructions organised in a unified dictionary-like format. His approach emphasises the integration of morphosyntactic information into constructional representations and considers constructions as the fundamental units for both language description and processing. Complementing these perspectives, Jakob Horsch (Slovak Academy of Sciences, Ľ. Štúr Institute of Linguistics) offered a Construction Grammar-based analysis of the Slovak comparative correlative, drawing on a corpus study to examine its formal structure, semantic patterns, and variation in clause and slot order. His findings highlight the role of meso-constructions and show how preferences such as iconic ordering shape the construction’s use, illustrating the descriptive precision that constructionist approaches can bring to the analysis of complex patterns.
Several talks addressed national and regional dictionary projects, reflecting the variety of lexicographic traditions represented at the conference. Contributions included plans for a new historical dictionary of Frisian, developments in the Termini Hungarian–Hungarian Dictionary and related educational terminology work, as well as papers from Croatia, Georgia, and Portugal describing the practical and methodological challenges facing contemporary dictionary-making. User-centred perspectives formed another recurrent theme. Papers by Yuliia Vasik & Teresa Fuentes (University of Salamanca) on dictionary design for visually impaired users and by Iasmin Valéria Miranda Rabelo et al. (Universidade Federal de Minas Gerais, Universidade Federal de Juiz de Fora,) on accessibility-related terminology drew attention to the need for lexicographic resources that accommodate diverse user groups and communicative contexts.
The poster session added further variety, ranging from corpus-based dialect research to OCR modelling for historical dictionaries and specialised lexicological studies. The selection included, among others, William Rolston Ashford’s (Dictionaries of the Scots Language) work on organising lexicographic evidence in the Dictionaries of the Scots Language and the analysis of oceanographic collocations by Sara de Piniés de la Cuesta and Javier Serrano Peinado (Universidad Complutense de Madrid).
The 2025 conference provided an opportunity for researchers working in lexicology, lexicography, corpus linguistics, and related areas to exchange ideas and discuss current methodological developments. The organisers intend to hold the event biennially, with the next conference planned for 2027. By sustaining this series, the Institute for Lexicology seeks to broaden professional networks, encourage methodological innovation, and strengthen both the Hungarian and international lexicographic communities. The participation and thematic variety of the 2025 programme indicate that LexiKonf can continue to serve as a valuable forum for colleagues interested in lexicographic and lexicological issues.