Elamur Research⌕
☰
home / papers / 2026 / A Corpus of Persuasion Techniques in Slavic Languages

A Corpus of Persuasion Techniques in Slavic Languages

Summary

Resources for detecting persuasion techniques in Slavic languages are scarce, especially outside news media. This paper fills that gap with a new annotated corpus covering Bulgarian and Polish parliamentary transcripts and Russian posts from community Telegram channels. The texts address contested issues such as the Ukraine-Russia war, defense, abortion law, EU and Schengen membership, and migration. Annotators labeled about 7,500 spans across 222 documents using the two-tier taxonomy from the SlavicNLP 2025 shared task, which has 25 techniques grouped into six strategies. Sentence-level labels were then derived by projecting the span annotations onto sentences. The authors report corpus statistics, analyze how topics relate to techniques, and benchmark classical ML models and LLM-based approaches. For sentence-level classification, fine-tuned RoBERTa-style models reach micro F1 of about .46 for Bulgarian, .44 for Polish and only .28 for Russian, with very low macro scores. These results show the task remains hard, particularly with little training data. The corpus and baseline models are freely released, giving researchers a benchmark for studying manipulative rhetoric in Slavic political and social-media discourse.

Key points

  • Coverage: 3 languages (BG, PL, RU) in 2 genres, with about 7,500 annotated spans in 222 documents. Labels exist at both span and sentence level, and the data forms part of the corpus behind the SlavicNLP 2025 shared task.
  • Taxonomy: the 25 fine-grained techniques are organized into 6 higher-level groups, namely Manipulative Wording, Call, Distraction, Simplification, Justification and Attack on Reputation.
  • Training data: the training splits contain roughly 3,138 sentences for Bulgarian, 2,809 for Polish and 1,101 for Russian. Because Russian had so little data, its model was trained on train+validation combined for 60 epochs.
  • Sentence-level results:
    • Micro F1 is .46 for BG, .44 for PL and .28 for RU; macro F1 is .24, .23 and .07.
    • The strongest classes are Name-Calling, Labeling (.68 BG), Appeal to Values and Appeal to Authority (.61 PL each), and Loaded Language (.59 PL).
    • Simplification and Distraction techniques mostly score near zero.
  • Models and release: the models include polish-roberta-8k, ruRoberta-large and xlm-roberta-large, and open LLMs were accessed through DeepInfra. Data is on GitHub (SlavicNLP/SlavicPersuasionTechniques) and models are in a Hugging Face SlavicNLP collection.

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

Fields: Linguistics & languages

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.63317/46rmmqosdrgt

@inproceedings{piskorski2026corpus,
  title = {{A Corpus of Persuasion Techniques in Slavic Languages}},
  author = {Piskorski, Jakub and Dimitrov, Dimitar Iliyanov and Ernst, Marina and Haneczok, Jacek and Marcinczuk, Michal and Modzelewski, Arkadiusz and Yangarber, Roman},
  booktitle = {Proceedings of the Fifteenth Language Resources and Evaluation Conference},
  year = {2026},
  doi = {10.63317/46rmmqosdrgt},
  url = {https://doi.org/10.63317/46rmmqosdrgt}
}

References

28 references, 4 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Abdurahmman Alzahrani, Eyad Babkier, Faisal Yanbaawi, Firas Yanbaawi, and Hassan Alhuzali. 2024. Investigating persuasion techniques in Arabic: An empirical study leveraging large language models. find our card
  2. Martin Atkinson, Jakub Piskorski, Erik van der Goot, and Roman Yangarber. 2011. Multilingual real-time event extraction for border security intelligence gathering. In U. Kock Wiil, editor, Counterterrorism and Open Source Intelligence, pages 355–390. Springer Lecture Notes in Social Networks, Vol. 2… find
  3. Alexis Conneau, Kartikay Khandelwal, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer, and Veselin Stoyanov. 2019. Unsupervised cross-lingual representation learning at scale. CoRR, abs/1911.02116. find
  4. Giovanni Da San Martino, Alberto Barrón-Cedeño, and Preslav Nakov. 2019a. Findings of the NLP4IF-2019 shared task on fine-grained propaganda detection. In Proceedings of the Second Workshop on Natural Language Processing for Internet Freedom: Censorship, Disinformation, and Propaganda, pages 162–170… find
  5. Giovanni Da San Martino, Alberto Barrón-Cedeño, Henning Wachsmuth, Rostislav Petrov, and Preslav Nakov. 2020. SemEval-2020 task 11: Detection of propaganda techniques in news articles. In Proceedings of the Fourteenth Workshop on Semantic Evaluation, pages 1377–1414, Barcelona (online). Internationa… find our card
  6. Giovanni Da San Martino, Seunghak Yu, Alberto Barrón-Cedeño, Rostislav Petrov, and Preslav Nakov. 2019b. Fine-grained analysis of propaganda in news articles. In EMNLP. find our card
  7. Dimitar Dimitrov, Firoj Alam, Maram Hasanain, Abul Hasnat, Fabrizio Silvestri, Preslav Nakov, and Giovanni Da San Martino. 2024. SemEval-2024 task 4: Multilingual detection of persuasion techniques in memes. In Proceedings of the 18th International Workshop on Semantic Evaluation (SemEval-2024), pag… find
  8. Dimitar Dimitrov, Bishr Bin Ali, Shaden Shaar, Firoj Alam, Fabrizio Silvestri, Hamed Firooz, Preslav Nakov, and Giovanni Da San Martino. 2021. SemEval-2021 task 6: Detection of persuasion techniques in texts and images. In Proceedings of the 15th International Workshop on Semantic Evaluation (SemEva… find
  9. Ivan Habernal, Raffael Hannemann, Christian Pollak, Christopher Klamm, Patrick Pauli, and Iryna Gurevych. 2017. Argotario: Computational argumentation meets serious games. In EMNLP. find
  10. Ivan Habernal, Patrick Pauli, and Iryna Gurevych. 2018. Adapting serious game for fallacious argumentation to German: Pitfalls, insights, and best practices. In LREC: Conference on Language Resources and Evaluation. find
  11. Silja Huttunen, Roman Yangarber, and Ralph Grishman. 2002. Diversity of scenarios in information extraction. In Proceedings of the Third International Conference on Language Resources and Evaluation (LREC 2002), Las Palmas de Gran Canaria, Spain. find
  12. Jan-Christoph Klie, Michael Bugert, Beto Boullosa, Richard Eckart de Castilho, and Iryna Gurevych. 2018. The INCEpTION platform: Machine-assisted and knowledge-oriented interactive annotation. In Proceedings of the 27th International Conference on Computational Linguistics: System Demonstrations, pa… find our card
  13. Roman Kyslyi, Nataliia Romanyshyn, and Volodymyr Sydorskyi. 2025. find
  14. The UNLP 2025 shared task on detecting social media manipulation. In Proceedings of the Fourth Ukrainian Natural Language Processing Workshop (UNLP 2025), pages 105–111, Vienna, Austria (online). Association for Computational Linguistics. find
  15. Arkadiusz Modzelewski, Giovanni Da San Martino, Pavel Savov, Magdalena Anna Wilczyńska, and Adam Wierzbicki. 2024. MIPD: Exploring manipulation and intention in a novel corpus of Polish disinformation. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages… find
  16. Arkadiusz Modzelewski, Witold Sosnowski, Tiziano Labruna, Adam Wierzbicki, and Giovanni Da San Martino. 2025. PCoT: Persuasion-augmented chain of thought for detecting fake news and social media disinformation. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistic… find
  17. Pablo Moral, Jesús M Fraile, Guillermo Marco, Anselmo Peñas, and Julio Gonzalo. 2024. Overview of dipromats 2024: Detection, characterization and tracking of propaganda in messages from diplomats and authorities of world powers. Procesamiento del lenguaje natural, 73:347–358. find
  18. Pablo Moral, Guillermo Marco, Julio Gonzalo, Jorge Carrillo-de Albornoz, and Iván Gonzalo-Verdugo. 2023. Overview of dipromats 2023: automatic detection and characterization of propaganda techniques in messages from diplomats and authorities of world powers. Procesamiento del lenguaje natural, 71:39… find
  19. Jakub Piskorski, Dimitar Iliyanov Dimitrov, Filip Dobranić, Marina Ernst, Jacek Haneczok, Ivan Koychev, Nikola Ljubešić, Michal Marcinczuk, Arkadiusz Modzelewski, Ivo Moravski, and Roman Yangarber. 2025. find
  20. SlavicNLP 2025 shared task: Detection and classification of persuasion techniques in parliamentary debates and social media. In Proceedings of the 10th Workshop on Slavic Natural Language Processing (Slavic NLP 2025), pages 254–275, Vienna, Austria. Association for Computational Linguistics. find
  21. Jakub Piskorski, Nicolas Stefanovitch, Firoj Alam, Ricardo Campos, Dimitar Dimitrov, Alípio Jorge, Senja Pollak, Nikolay Ribin, Zoran Fijavz, Maram Hasanain, Purificação Silvano, Elisa Sartori, Nuno Guimarães, Ana Zwitter Vitez, Ana Filipa Pacheco, Ivan Koychev, Nana Yu, Preslav Nakov, and Giovanni… find
  22. Jakub Piskorski, Nicolas Stefanovitch, Valerie-Anne Bausier, Nicolo Faggiani, Jens Linge, Sopho Kharazi, Nikolaos Nikolaidis, Giulia Teodori, Bertrand De Longueville, Brian Doherty, Jason Gonin, Camelia Ignat, Bonka Kotseva, Eleonora Mantica, Lorena Marcaletti, Enrico Rossi, Alessio Spadaro, Marco V… find
  23. Jakub Piskorski, Nicolas Stefanovitch, Giovanni Da San Martino, and Preslav Nakov. 2023b. SemEval-2023 task 3: Detecting the category, the framing, and the persuasion techniques in online news in a multi-lingual setup. In Proceedings of the 17th International Workshop on Semantic Evaluation (SemEval… find
  24. Jakub Piskorski, Nicolas Stefanovitch, Nikolaos Nikolaidis, Giovanni Da San Martino, and Preslav Nakov. 2023c. Multilingual multifaceted understanding of online news in terms of genre, framing, and persuasion techniques. In Proceedings of the 61st Annual Meeting of the Association for Computational… find
  25. Jakub Piskorski and Roman Yangarber. 2013. Information extraction: past, present and future. In Multi-source, multilingual information extraction and summarization, pages 23–49. Springer. find
  26. Nicolas Stefanovitch and Jakub Piskorski. 2023. Holistic inter-annotator agreement and corpus coherence estimation in a large-scale multilingual annotation campaign. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 71–86, Singapore. Association for Com… find
  27. Roman Yangarber. 2006. Verification of facts across document boundaries. In Proceedings of the International Workshop on Intelligent Information Access (IIIA-2006), Helsinki, Finland. find
  28. Dmitry Zmitrovich, Aleksandr Abramov, Andrey Kalmykov, Vitaly Kadulin, Maria Tikhonova, Ekaterina Taktasheva, Danil Astafurov, Mark Baushenko, Artem Snegirev, Tatiana Shavrina, Sergei S. Markov, Vladislav Mikhailov, and Alena Fenogenova. 2024. A family of pre-trained transformer language models for… find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/piskorski-2026-corpus-persuasion-techniques-slavic-languages/