Large Language Models for Word-In-Context
Main Article Content
Abstract
Task-specific models have long dominated natural language processing tasks. However, general-purpose large language models (LLMs) have recently begun to successfully compete with highly specialized solutions across various NLP domains. In this paper, we investigate the applicability of LLMs to the task of estimating the semantic proximity of a word's meanings in a pair of usages, known as Word-in-Context (WiC). Drawing on the multilingual CoMeDi benchmark, we propose novel approaches to building automated WiC-systems based on LLMs. We conduct a systematic comparison of five different configurations in terms of quality and computational costs. In particular, we propose a configuration where LLM predictions are adjusted using a training set, without the need to fine-tune the LLM itself. Results on test sets across seven languages show that our proposed approaches enable LLMs to outperform all existing specialized systems, establishing a new state-of-the-art (SOTA) on the CoMeDi benchmark. Nevertheless, the achieved high quality comes with a significant increase in computational costs: LLM-based systems require several orders of magnitude more computations compared to compact specialized models (such as XL-DURel). This work represents a step towards understanding the trade-off between accuracy and resource efficiency when using modern LLMs in lexical semantics tasks.
Article Details
References
2. Tahmasebi N., Borin L., Jatowt A., Xu Y., Hengchen S. Computational Approaches to Semantic Change. Berlin: Language Science Press, 2021.https://doi.org/10.5281/zenodo.5040241
3. Hendy A. et al. How Good Are GPT Models at Machine Translation? A Comprehensive Evaluation // arXiv:2302.09210 [cs.CL], 2023. https://doi.org/10.48550/arXiv.2302.09210
4. Liang P. et al. Holistic Evaluation of Language Models // arXiv: 2211.09110 [cs.CL], 2022. https://doi.org/10.48550/arXiv.2211.09110
5. Schlechtweg D., Schulte im Walde S., Eckmann S. Diachronic Usage Relatedness (DURel): A Framework for the Annotation of Lexical Semantic Change // Proc. of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Vol. 2. New Orleans, Louisiana, June 2018. Association for Computational Linguistics, 2018. P. 169–174. https://doi.org/10.18653/v1/N18-2027.
6. Arefyev N., Fedoseev M., Protasov V., Homskiy D., Davletov A., Panchenko A. DeepMistake: Which Senses are Hard to Distinguish for a Word-in-Context Model // Computational Linguistics and Intellectual Technologies (Dialog 2021). Russian Federation, 2021. 20. P. 16–30.
7. Cassotti P., Siciliani L., DeGemmis M., Semeraro G., Basile P. XL-LEXEME: WiC Pretrained Model for Cross-Lingual LEXical sEMantic changE // Proc. of the 61st Annual Meeting of the Association for Computational Linguistics. Vol. 2. Toronto, Canada, July 2023. Association for Computational Linguistics, 2023. С. 1577–1585. https://doi.org/10.18653/v1/2023.acl-short.135
8. Yadav S., Schlechtweg D. XL-DURel: Finetuning Sentence Transformers for Ordinal Word-in-Context Classification // arXiv:2507.14578 [cs.CL], 2025.https://doi.org/10.48550/arXiv.2507.14578
9. Devlin J., Chang M.-W., Lee K., Toutanova K. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding // arXiv:1810.04805 [cs.CL], 2018. https://doi.org/10.48550/arXiv.1810.04805
10. Conneau A. et al. Unsupervised Cross-lingual Representation Learning at Scale // arXiv:1911.02116 [cs.CL], 2019. https://doi.org/10.48550/arXiv.1911.02116
11. Loke Y. X., Schlechtweg D., Zhao W. ABDN-NLP at CoMeDi Shared Task: Predicting the Aggregated Human Judgment via Weighted Few-Shot Prompting // Proc. of Context and Meaning: Navigating Disagreements in NLP Annotation. Abu Dhabi, UAE, January 2025. International Committee on Computational Linguistics, 2025. P. 122–128.
12. Periti F., Tahmasebi N. A Systematic Comparison of Contextualized Word Embeddings for Lexical Semantic Change // Proc. of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Vol. 1. Mexico City, Mexico, June 2024. Association for Computational Linguistics, 2024. P. 4262–4282. https://doi.org/10.18653/v1/2024.naacl-long.240
13. Yadav S., Choppa T., Schlechtweg D. Towards Automating Text Annotation: A Case Study on Semantic Proximity Annotation using GPT-4 // arXiv:2407.04130 [cs.CL], 2024. https://doi.org/10.48550/arXiv.2407.04130
14. Zamora-Reina F. D., Bravo-Marquez F., Schlechtweg D., Arefyev N. Can Large Language Models Compete with Specialized Models in Lexical Semantic Change Detection? // ECAI 2025. Frontiers in Artificial Intelligence and Applications. Vol. 413. IOS Press, 2025. P. 4201–4208. https://doi.org/10.3233/FAIA251313.
15. Schlechtweg D., Choppa T., Zhao W., Roth M. CoMeDi Shared Task: Median Judgment Classification & Mean Disagreement Ranking with Ordinal Word-in-Context Judgments // Proc. of Context and Meaning: Navigating Disagreements in NLP Annotation. Abu Dhabi, UAE, January 2025. International Committee on Computational Linguistics, 2025. P. 33–47.
16. Yang A. et al. Qwen3 Technical Report // arXiv:2505.09388 [cs.CL], 2025. https://doi.org/10.48550/arXiv.2505.09388
17. Krippendorff K. Content Analysis: An Introduction to Its Methodology. 4th ed. Thousand Oaks, CA: SAGE Publications, 2018.
18. Sakai T. Evaluating Evaluation Measures for Ordinal Classification and Ordinal Quantification // Proc. of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing. Vol. 1. Online, August 2021. Association for Computational Linguistics, 2021. P. 175–187. https://doi.org/10.18653/v1/2021.acl-long.15
19. Kuklin M., Arefyev N. Deep-change at CoMeDi: The Cross-Entropy Loss is not All You Need // Proc. of Context and Meaning: Navigating Disagreements in NLP Annotation. Abu Dhabi, UAE, January 2025. International Committee on Computational Linguistics, 2025. P. 48–64.
20. Li X., Li J. AnglE-optimized Text Embeddings // arXiv: 2309.12871. 2023. https://doi.org/10.48550/arXiv.2309.12871
21. Schlechtweg D., Tahmasebi N., Hengchen S., Dubossarsky H., McGillivray B. DWUG: A large Resource of Diachronic Word Usage Graphs in Four Languages // Proc. of the 2021 Conference on Empirical Methods in Natural Language Processing. Online and Punta Cana, Dominican Republic, November 2021. Association for Computational Linguistics, 2021. P. 7079–7091. https://doi.org/10.18653/v1/2021.emnlp-main.567
22. Khattab O. et al. DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines // arXiv: 2310.03714 [cs.CL], 2023. https://doi.org/10.48550/arXiv.2310.03714
23. Shao Z. et al. DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models // arXiv:2402.03300 [cs.CL], 2024. https://doi.org/10.48550/arXiv.2402.03300
24. DeepSeek-AI, Guo D. et al. DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning // arXiv:2501.12948 [cs.CL]. https://doi.org/10.1038/s41586-025-09422-z
25. Woosuk K. et al. Efficient Memory Management for Large Language Model Serving with PagedAttention // Proc. of the ACM SIGOPS 29th Symposium on Operating Systems Principles. 2023. https://doi.org/10.1145/3600006.3613165
This work is licensed under a Creative Commons Attribution 4.0 International License.
Presenting an article for publication in the Russian Digital Libraries Journal (RDLJ), the authors automatically give consent to grant a limited license to use the materials of the Kazan (Volga) Federal University (KFU) (of course, only if the article is accepted for publication). This means that KFU has the right to publish an article in the next issue of the journal (on the website or in printed form), as well as to reprint this article in the archives of RDLJ CDs or to include in a particular information system or database, produced by KFU.
All copyrighted materials are placed in RDLJ with the consent of the authors. In the event that any of the authors have objected to its publication of materials on this site, the material can be removed, subject to notification to the Editor in writing.
Documents published in RDLJ are protected by copyright and all rights are reserved by the authors. Authors independently monitor compliance with their rights to reproduce or translate their papers published in the journal. If the material is published in RDLJ, reprinted with permission by another publisher or translated into another language, a reference to the original publication.
By submitting an article for publication in RDLJ, authors should take into account that the publication on the Internet, on the one hand, provide unique opportunities for access to their content, but on the other hand, are a new form of information exchange in the global information society where authors and publishers is not always provided with protection against unauthorized copying or other use of materials protected by copyright.
RDLJ is copyrighted. When using materials from the log must indicate the URL: index.phtml page = elbib / rus / journal?. Any change, addition or editing of the author's text are not allowed. Copying individual fragments of articles from the journal is allowed for distribute, remix, adapt, and build upon article, even commercially, as long as they credit that article for the original creation.
Request for the right to reproduce or use any of the materials published in RDLJ should be addressed to the Editor-in-Chief A.M. Elizarov at the following address: amelizarov@gmail.com.
The publishers of RDLJ is not responsible for the view, set out in the published opinion articles.
We suggest the authors of articles downloaded from this page, sign it and send it to the journal publisher's address by e-mail scan copyright agreements on the transfer of non-exclusive rights to use the work.