Elamur Research⌕
☰
home / papers / 2026 / Elections go bananas: A First Large-scale Multilingual Study of Pluralia Tantum using LLMs

Elections go bananas: A First Large-scale Multilingual Study of Pluralia Tantum using LLMs

Paper · 2026 · Elena Spaziani, Kamyar Zeinalipour, Pierluigi Cassotti et al.

Summary

The paper examines pluralia tantum, nouns whose plural form dominates or fully replaces the singular (e.g., scissors, RU выборы 'elections'). It asks how such nouns develop over time, and in particular whether full loss of the singular (morphological pluralia tantum, mPT) grows out of plural-dominant (PD) nouns or semantic pluralia tantum (sPT). The authors gathered candidate nouns for Italian, Russian and English from one reference dictionary per language. They kept only words whose singular/plural ratio visibly shifts in a diachronic corpus. OpenAI and DeepSeek models were then tested on a hand-annotated set using the Lexicalization profiles framework. After the models proved adequate, DeepSeek Reasoner annotated every corpus context for semantic, syntactic and sense features. These large-scale annotations show the plural becoming semantically separated from the singular. Semantic features turn out more informative than colligational ones. Case studies of fondello and rifiuto suggest that plural senses can move toward either mPT or sPT. This is presented as the first computational, cross-linguistic, multi-century study of the phenomenon.

Key points

  • Data sources by language:
    • Italian: De Mauro's dictionary and the La Stampa Corpus.
    • Russian: Ožegov & Švedova's dictionary and the Russian National Corpus. Pre-reform spelling was converted to modern orthography, and UDPipe and Pymorphy were used for lemmatization.
    • English: the OED and The Times Archive.
  • Word selection: dictionaries yielded roughly 800–5,000 candidates per language. Frequency filtering reduced these to 70 (IT), 60 (RU) and 241 (EN) target words.
  • Model accuracy: on the hand-annotated test set, LLM accuracy averaged across feature groups and languages ranged from 51% to 89%. Sense relations remained the hardest category.
  • Annotation scale: DeepSeek Reasoner processed about 1.5M prompts, with an average of 1,592, 978 and 1,284 contexts per word for IT, RU and EN.
  • Limitations: three-shot prompting slightly helped overall but lowered Italian scores, so the authors judge zero-shot prompting the safer choice. Corpus licensing prevents release of the annotated contexts, although the code is shared.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓Sapienza University of RomeRome · Italy57 works · 9,564 citations🎓University of GothenburgGothenburg · Sweden34 works · 1,625 citations🎓University of SienaSiena · Italy9 works · 198 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 works

🧭 Topics 1Research subjects and languages the work is about.Multilingual & cross-lingual NLP · 828 works

Cite

Cite as Elena Spaziani, Kamyar Zeinalipour, Pierluigi Cassotti and Nina Tahmasebi. 2026. Elections go bananas: A First Large-scale Multilingual Study of Pluralia Tantum using LLMs. Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.eacl-long.308

@inproceedings{spaziani2026elections,
  title = {{Elections go bananas: A First Large-scale Multilingual Study of Pluralia Tantum using LLMs}},
  author = {Elena Spaziani, Kamyar Zeinalipour, Pierluigi Cassotti and Nina Tahmasebi},
  booktitle = {Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.eacl-long.308},
  url = {https://doi.org/10.18653/v1/2026.eacl-long.308}
}

References

33 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Keith Allan. 1980. Nouns and countability. Language. find
  2. Andreas Blank. 1997. Prinzipien des lexikalischen Bedeutungswandels am Beispiel der romanischen Sprachen. Max Niemeyer Verlag. find
  3. Leonard Bloomfield. 1984. Language. University of Chicago Press. find
  4. Geert Booij. 2012. The Grammar of Words. An Introduction to Linguistic Morphology. Oxford University Press. find
  5. Maximilian Braun. 1930. Das Kollektivum und das Plurale tantum im Russischen: Ein bedeutungs-geschichtlicher Versuch. Ph.D. thesis, Universitat Leipzig. find
  6. Bernard Comrie. 1981. Language Universals and Linguistic Typology. Blackwell. find
  7. Greville Corbett. 2000. Number. Cambridge University Press. find
  8. Greville G. Corbett. 2018. Pluralia tantum nouns and the theory of features: a typology of nouns with non-canonical number properties. Morphology. find
  9. Tullio De Mauro. 2007. Grande dizionario italiano dell’uso. UTET. find
  10. Vladimir I. Degtjarev. 2014. Kategorija čisla v slavjan-skich jazykach (istoriko-semantičeskoe issledovanie. Izdatel’stvo Južnogo Federal’nogo Universiteta. find
  11. Dagmar Divjak and Stefan Th. Gries. 2006. Ways of trying in russian: Clustering behavioral profiles. Corpus Linguistics and Linguistic Theory. find
  12. Laura Gardelle and Paolo Acquaviva. 2023. Pluralia tantum and singularia tantum. The Wiley Blackwell Companion to Morphology. find
  13. Joseph H. Greenberg. 1963. Universals of language. M. I. T. Press. find
  14. Stefan Th. Gries. 2010. Behavioral profiles: A fine-grained and quantitative approach in corpus-based lexical semantics. The Mental Lexicon. find
  15. Martin Haspelmath and Andres Karjus. 2017. Explaining asymmetries in number marking: Singulatives, pluratives and usage frequency. Linguistics. find
  16. Raihan Kibria, Sheikh Intiser Uddin Dipta, and Muhammad Abdullah Adnan. 2024. On functional competence of LLMs for linguistic disambiguation. In Proceedings of the 28th Conference on Computational Natural Language Learning, pages 143–160, Miami, FL, USA. Association for Computational Linguistics. find
  17. R. Killick, P. Fearnhead, and I. A. Eckley and. 2012. Optimal detection of changepoints with a linear computational cost. Journal of the American Statistical Association, 107(500):1590–1598. find
  18. Peter Koch. 2001. Metonymy: Unity in diversity. Journal of Historical Pragmatics. find
  19. Maria Koptjevskaja-Tamm and B. Wälchli. 2001. The circum-baltic languages: An areal-typological approach. In The Circum-Baltic Languages: Typology and Contact. Vol. 2. John Benjamins Publishing. find
  20. Mikhail Korobov. 2015. Morphological analyzer and generator for russian and ukrainian languages. In Analysis of Images, Social Networks and Texts. Springer International Publishing. find
  21. Andrey Kutuzov, Lidia Pivovarova, and Mario Giulianelli. 2021. Grammatical profiling for semantic change detection. In Proceedings of the 25th Conference on Computational Natural Language Learning, pages 423–434, Online. Association for Computational Linguistics. find
  22. Peter Hugoe Matthews. 2007. The Concise Oxford Dictionary of Linguistics. Oford University Press. find
  23. Oxford English Dictionary OED. 1989. Oxford english dictionary. Simpson, Ja & Weiner, Esc, 3. find
  24. Sergej Ožegov and Natalija Švedova. 1992. Tol’kovyj slovar’ russkogo jazyka. Izdatel’stvo Az’. find
  25. Edward Sapir. 1921. Language, an introduction to the study of speech. Harcourt, Brace and Company. find
  26. Elena Spaziani. Forthcoming. Nizy, koški, peregovory and vybory: Lexicalization profiles. Studi Slavistici. find
  27. Milan Straka. 2018. UDPipe 2.0 prototype at CoNLL 2018 UD shared task. In Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, pages 197–207, Brussels, Belgium. ACL.
  28. Anna Thornton. 2004. La formazione delle parole in italiano, chapter Conversione. Max Niemeyer Verlag. find
  29. Charles Truong, Laurent Oudre, and Nicolas Vayatis. 2020. Selective review of offline change point detection methods. Signal Processing, 167:107299. find
  30. Leonie Weissweiler, Abdullatif Köksal, and Hinrich Schütze. 2024. Hybrid human-llm corpus construction and llm evaluation for rare linguistic phenomena. arXiv preprint arXiv:2403.06965.
  31. Martina Wiltschko. 2008. The syntax of non-inflectional plural marking. Nat Lang Linguistic Theory. find
  32. Elena Šemaeva. 2016. Identificirujuščaja metaforičeskaja model’ leksikalizacii form množestvennogo čisla anglijskich i russkich imen suščestvitel’nych. In Kognitivnye issledovanija jazyka. Izdatel’skij dom TGU im. G.R. Deržavina. find
  33. Elena V. Šemaeva. 2014. Kognitivnye osnovy leksikalizacii form množestvennogo čisla imeni suščestvitel’nogo: na materiale anglijskogo i russkogo jazykov. Ph.D. thesis, Belgorodskij Gosu-darstvennyj Nacionalnyj Issledovatel’skij Universitet. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/spaziani-2026-elections-go-bananas-first-large/