Elamur Research⌕
☰
home / papers / 2026 / Multilingual and Cross-Lingual Citation Needed Detection on Wikipedia for Lower-Resource Languages

Multilingual and Cross-Lingual Citation Needed Detection on Wikipedia for Lower-Resource Languages

Paper · 2026 · Gerrit Quaremba, Amy Rechkemmer, Elizabeth Black et al.

Summary

Wikipedia editors mark unsupported statements with a "citation needed" tag. Automating this step, called Citation Needed Detection (CND), has mostly been studied for large Wikipedias, and recent fact-checking pipelines tend to rely on costly proprietary LLMs. The authors build MCN, a corpus of sentences drawn from Featured Articles in 18 languages, including Ukrainian. Each sentence is labelled by whether it carries an inline citation. They adapt open decoder models of roughly 8B parameters (SLMs) with QLoRA under several supervised fine-tuning variants. These are compared with encoder PLMs and prompted LLMs, both per language and cross-lingually, where models are trained on English alone. An encoder-style training objective gives the best SLMs, and fine-tuned compact models beat prompted LLMs at every resource level. English-only models transfer well zero-shot on translated parallel data but lose about 20 points on genuine target-language claims, while still outperforming LLMs. For under-staffed Wikipedia communities, this points to small, task-specific models as an affordable option.

Key points

  • MCN covers 18 languages: 8 high-, 5 medium- and 5 low-resource (Joshi et al. taxonomy). Per-language sizes range from about 2K to 1.1m claims. Ukrainian is a medium-resource language, with 1,343 Featured Articles.
  • SLMs are fine-tuned with QLoRA. Encoder-style fine-tuning (Suganthan et al., 2025) improves on generative SFT variants by up to 8 percentage points.
  • Fine-tuned SLMs and encoder PLMs consistently outperform prompted LLMs, with PLMs remaining a strong baseline. Performance still varies widely across resource levels (roughly 53–75%).
  • In the cross-lingual setting (X-CND), English-trained models drop by up to ~20 points under real-world conditions compared with parallel data. Even so, ES-tuned SLMs beat LLMs with zero or minimal target-language data. The findings also hold on randomly sampled Wikipedia claims, not just Featured Articles.
  • Data, including the Google Translate–based parallel text, is released under CC BY-SA, and code is available at github.com/gerritq/mcn. Stated limitations:
    • 70% of the languages are Indo-European.
    • The labels follow a citation-presence heuristic.
    • Results are restricted to the Wikipedia domain.

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

🎓King's College LondonLondon · United Kingdom42 works · 1,617 citations🤝Wikimedia FoundationSan Francisco · United States2 works · 255 citations

 Fields, topics and concepts

📚 Shelves 6The 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 worksSocial Sciences · 1,772 worksSociology and Political Science · 577 works

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

Cite

Cite as Gerrit Quaremba, Amy Rechkemmer, Elizabeth Black, Denny Vrandečić and Elena Simperl. 2026. Multilingual and Cross-Lingual Citation Needed Detection on Wikipedia for Lower-Resource Languages. Proceedings of the 64th Annual Meeting 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.acl-long.1291

@inproceedings{quaremba2026multilingual,
  title = {{Multilingual and Cross-Lingual Citation Needed Detection on Wikipedia for Lower-Resource Languages}},
  author = {Gerrit Quaremba, Amy Rechkemmer, Elizabeth Black, Denny Vrandečić and Elena Simperl},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.1291},
  url = {https://doi.org/10.18653/v1/2026.acl-long.1291}
}

References

53 references, 19 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Fatma Arslan, Naeemul Hassan, Chengkai Li, and Mark Tremayne. 2020. A benchmark dataset of check-worthy factual claims. In Proceedings of the international AAAI conference on web and social media, volume 14, pages 821–829. find
  2. Mikel Artetxe, Gorka Labaka, and Eneko Agirre. 2020. Translation artifacts in cross-lingual transfer learning. arXiv preprint arXiv:2004.04721.
  3. Aitolkyn Baigutanova, Jaehyeon Myung, Diego Saez-Trumper, Ai-Jou Chou, Miriam Redi, Changwook Jung, and Meeyoung Cha. 2023. Longitudinal assessment of reference quality on wikipedia. In Proceedings of the ACM web conference 2023, pages 2831–2839. find
  4. Aitolkyn Baigutanova, Francisco Navas, Pablo Aragón, Mykola Trokhymovych, Muniza Aslam, Ai-Jou Chou, Miriam Redi, and Diego Sáez Trumper. 2026. Multilingual reference need assessment system for wikipedia. In Proceedings of the ACM Web Conference 2026, WWW ’26, page 8191–8199, New York, NY, USA. Asso… find
  5. Peter Belcak, Greg Heinrich, Shizhe Diao, Yonggan Fu, Xin Dong, Saurav Muralidharan, Yingyan Celine Lin, and Pavlo Molchanov. 2025. Small language models are the future of agentic ai. Preprint, arXiv:2506.02153. find
  6. Andrew Bell. 2025. Less can be more: An empirical evaluation of small and large language models for sentence-level claim detection. In Proceedings of the Eighth Fact Extraction and VERification Workshop (FEVER), pages 85–90. find
  7. Elliot Bolton, Abhinav Venigalla, Michihiro Yasunaga, David Hall, Betty Xiong, Tony Lee, Roxana Daneshjou, Jonathan Frankle, Percy Liang, Michael Carbin, and 1 others. 2024. Biomedlm: A 2.7 b parameter language model trained on biomedical text. arXiv preprint arXiv:2403.18421.
  8. Martin Juan José Bucher and Marco Martini. 2024. Fine-tuned’small’llms (still) significantly outperform zero-shot generative ai models in text classification. arXiv preprint arXiv:2406.08660. — [arXiv:2406.08660](https://arxiv.org/abs/2406.08660) find
  9. Eirini Chatzikoumi. 2020. How to evaluate machine translation: A review of automated and human metrics. Natural Language Engineering, 26(2):137–161. find
  10. Gheorghe Comanici, Eric Bieber, Mike Schaekermann, Ice Pasupat, Noveen Sachdeva, Inderjit Dhillon, Marcel Blistein, Ori Ram, Dan Zhang, Evan Rosen, Luke Marris, Sam Petulla, Colin Gaffney, Asaf Aharoni, Nathan Lintz, Tiago Cardal Pais, Henrik Jacobsson, Idan Szpektor, Nan-Jiang Jiang, and 3416 other…
  11. Alexis Conneau, Kartikay Khandelwal, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer, and Veselin Stoyanov. 2020. Unsupervised cross-lingual representation learning at scale. Preprint, arXiv:1911.02116. find
  12. Aida Halitaj and Arkaitz Zubiaga. 2025. Alpet: Active few-shot learning for citation worthiness detection in low-resource wikipedia languages. Expert Systems With Applications, 281:127503. find
  13. Tim Dettmers, Artidoro Pagnoni, Ari Holtzman, and Luke Zettlemoyer. 2023. Qlora: Efficient finetuning of quantized llms. Preprint, arXiv:2305.14314.
  14. Maram Hasanain, Reem Suwaileh, Sanne Weering, Chengkai Li, Tommaso Caselli, Wajdi Zaghouani, Alberto Barrón-Cedeño, Preslav Nakov, and Firoj Alam. 2024. Overview of the clef-2024 checkthat! lab task 1 on check-worthiness estimation of multigenre content. In Working Notes of the Conference and Labs o… find
  15. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. Bert: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human language technologies,…
  16. Alphaeus Dmonte, Roland Oruche, Marcos Zampieri, Prasad Calyam, and Isabelle Augenstein. 2024. Claim verification in the age of large language models: A survey. arXiv preprint arXiv:2408.14317.
  17. Subhabrata Dutta, Rudra Dhar, Prantik Guha, Arpan Murmu, and Dipankar Das. 2023. A multilingual dataset for identification of factual claims in indian twitter. In Proceedings of the 14th Annual Meeting of the Forum for Information Retrieval Evaluation, FIRE ’22, page 88–92, New York, NY, USA. Associ… find
  18. Revanth Gangi Reddy, Sai Chetan Chinthakindi, Zhenhailong Wang, Yi Fung, Kathryn Conger, Ahmed ELsayed, Martha Palmer, Preslav Nakov, Eduard Hovy, Kevin Small, and Heng Ji. 2022. NewsClaims: A new benchmark for claim detection from news with attribute knowledge. In Proceedings of the 2022 Conference… find
  19. Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Alex Vaughan, Amy Yang, Angela Fan, Anirudh Goyal, Anthony Hartshorn, Aobo Yang, Archi Mitra, Archie Sravankumar, Artem Korenev, Arthur Hinsvark, and 542 o… — The Llama 3 Herd of Models
  20. Zhijiang Guo, Michael Schlichtkrull, and Andreas Vlachos. 2022. A survey on automated fact-checking. Transactions of the association for computational linguistics, 10:178–206.
  21. Naeemul Hassan, Bill Adair, James T Hamilton, Chengkai Li, Mark Tremayne, Jun Yang, and Cong Yu. 2015. The quest to automate fact-checking. In Proceedings of the 2015 computation+ journalism symposium. The Computation+ Journalism Symposium New York. find
  22. Isaac Johnson, Lucie-Aimée Kaffee, and Miriam Redi. 2024. Wikimedia data for ai: a review of wikimedia datasets for nlp tasks and ai-assisted editing. arXiv preprint arXiv:2410.08918. find
  23. Pratik Joshi, Sebastin Santy, Amar Budhiraja, Kalika Bali, and Monojit Choudhury. 2020. The state and fate of linguistic diversity and inclusion in the nlp world. arXiv preprint arXiv:2004.09095.
  24. Ashkan Kazemi, Kiran Garimella, Devin Gaffney, and Scott A. Hale. 2021. Claim matching beyond English to scale global fact-checking. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (… find
  25. Lev Konstantinovskiy, Oliver Price, Mevan Babakar, and Arkaitz Zubiaga. 2021. Toward automated factchecking: Developing an annotation schema and benchmark for consistent automated claim detection. Digital threats: research and practice, 2(2):1–16. find
  26. Aaron Halfaker and R. Stuart Geiger. 2020. Ores: Lowering barriers with participatory machine learning in wikipedia. Preprint, arXiv:1909.05189. — [arXiv:1909.05189](https://arxiv.org/abs/1909.05189) find
  27. Anne Lauscher, Vinit Ravishankar, Ivan Vulić, and Goran Glavaš. 2020. From zero to hero: On the limitations of zero-shot language transfer with multilingual Transformers. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 4483–4499, Online. Asso…
  28. Aida Halitaj and Arkaitz Zubiaga. 2024. Providing citations to support fact-checking: Contextualizing detection of sentences needing citation on small wikipedias. Natural Language Processing Journal, 8:100093. find
  29. Włodzimierz Lewoniewski, Krzysztof W˛ecel, and Witold Abramowicz. 2017. Analysis of references across wikipedia languages. In International Conference on Information and Software Technologies, pages 561–573. Springer. find
  30. Lujun Li, Lama Sleem, Niccolo’ Gentile, Geoffrey Nichil, and Radu State. 2025. Small language models in the real world: Insights from industrial text classification. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track), pages 971–982,… find
  31. Miaoran Li, Baolin Peng, Michel Galley, Jianfeng Gao, and Zhu Zhang. 2024a. Self-checker: Plug-and-play modules for fact-checking with large language models. In Findings of the Association for Computational Linguistics: NAACL 2024, pages 163–181.
  32. Yufeng Li, Rrubaa Panchendrarajan, and Arkaitz Zubiaga. 2024b. Factfinders at checkthat! 2024: refining check-worthy statement detection with llms through data pruning. arXiv preprint arXiv:2406.18297. — [arXiv:2406.18297](https://arxiv.org/abs/2406.18297) find
  33. Chin-Yew Lin. 2004. ROUGE: A package for automatic evaluation of summaries. In Text Summarization Branches Out, pages 74–81, Barcelona, Spain. Association for Computational Linguistics.
  34. Laura Majer and Jan Šnajder. 2024. Claim check-worthiness detection: How well do llms grasp annotation guidelines? arXiv preprint arXiv:2404.12174. find
  35. Shervin Minaee, Tomas Mikolov, Narjes Nikzad, Meysam Chenaghlu, Richard Socher, Xavier Amatriain, and Jianfeng Gao. 2025. Large language models: A survey. Preprint, arXiv:2402.06196.
  36. Preslav Nakov, David Corney, Maram Hasanain, Firoj Alam, Tamer Elsayed, Alberto Barrón-Cedeño, Paolo Papotti, Shaden Shaar, and Giovanni Da San Martino. 2021. Automated fact-checking for assisting human fact-checkers. arXiv preprint arXiv:2103.07769. — [arXiv:2103.07769](https://arxiv.org/abs/2103.07769) find
  37. OpenAI. 2025. Gpt-5 system card. https://cdn.openai.com/gpt-5-system-card.pdf. Accessed: 20225-12-12. ↗
  38. Rrubaa Panchendrarajan and Arkaitz Zubiaga. 2024. Claim detection for automated fact-checking: A survey on monolingual, multilingual and cross-lingual research. Natural Language Processing Journal, 7:100066. find
  39. Subhadarshi Panda and Sarah Ita Levitan. 2021. Detecting multilingual covid-19 misinformation on social media via contextualized embeddings. In Proceedings of the Fourth Workshop on NLP for Internet Freedom: Censorship, Disinformation, and Propaganda, pages 125–129.
  40. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. Bleu: a method for automatic evaluation of machine translation. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, pages 311–318, Philadelphia, Pennsylvania, USA. Association for Computational…
  41. Miriam Redi, Besnik Fetahu, Jonathan Morgan, and Dario Taraborelli. 2019. Citation needed: A taxonomy and algorithmic assessment of wikipedia’s verifiability. In The world wide web conference, pages 1567–1578. find
  42. Shaden Shaar, Firoj Alam, Giovanni Da San Martino, Alex Nikolov, Wajdi Zaghouani, Preslav Nakov, and Anna Feldman. 2021. Findings of the NLP4IF-2021 shared tasks on fighting the COVID-19 infodemic and censorship detection. In Proceedings of the Fourth Workshop on NLP for Internet Freedom: Censorship… find
  43. Paul Suganthan, Fedor Moiseev, Le Yan, Junru Wu, Jianmo Ni, Jay Han, Imed Zitouni, Enrique Alfonseca, Xuanhui Wang, and Zhe Dong. 2025. Adapting decoder-based language models for diverse encoder downstream tasks. arXiv preprint arXiv:2503.02656. — [arXiv:2503.02656](https://arxiv.org/abs/2503.02656) find
  44. Ahmet Üstün, Viraat Aryabumi, Zheng Yong, Wei-Yin Ko, Daniel D’souza, Gbemileke Onilude, Neel Bhandari, Shivalika Singh, Hui-Lee Ooi, Amr Kayid, and 1 others. 2024. Aya model: An instruction finetuned open-access multilingual language model. In Proceedings of the 62nd Annual Meeting of the Associati…
  45. Andreas Vlachos and Sebastian Riedel. 2014. Fact checking: Task definition and dataset construction. In Proceedings of the ACL 2014 workshop on language technologies and computational social science, pages 18–22.
  46. Leandro von Werra, Younes Belkada, Lewis Tunstall, Edward Beeching, Tristan Thrush, Nathan Lambert, Shengyi Huang, Kashif Rasul, and Quentin Gallouédec. 2020. Trl: Transformer reinforcement learning. https://github.com/huggingface/trl. ↗
  47. Yuxia Wang, Revanth Gangi Reddy, Zain Muhammad Mujahid, Arnav Arora, Aleksandr Rubashevskii, Jiahui Geng, Osama Mohammed Afzal, Liangming Pan, Nadav Borenstein, Aditya Pillai, Isabelle Augenstein, Iryna Gurevych, and Preslav Nakov. 2024. Factcheck-bench: Fine-grained evaluation benchmark for automat… find
  48. Greta Warren, Irina Shklovski, and Isabelle Augenstein. 2025. Show me the work: Fact-checkers’ requirements for explainable automated fact-checking. In Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems, pages 1–21. find
  49. Jason Wei, Maarten Bosma, Vincent Y Zhao, Kelvin Guu, Adams Wei Yu, Brian Lester, Nan Du, Andrew M Dai, and Quoc V Le. 2021. Finetuned language models are zero-shot learners. arXiv preprint arXiv:2109.01652.
  50. Jillian Xu, Dylan Zhou, Vinay Shukla, Yang Yang, Junrui Ruan, Shuhuai Lin, Wenfei Zou, Yinxiao Liu, and Karthik Lakshmanan. 2025. Dual-head reasoning distillation: Improving classifier accuracy with train-time-only reasoning. Preprint, arXiv:2509.21487. — [arXiv:2509.21487](https://arxiv.org/abs/2509.21487) find
  51. An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, Chujie Zheng, Dayiheng Liu, Fan Zhou, Fei Huang, Feng Hu, Hao Ge, Haoran Wei, Huan Lin, Jialong Tang, and 41 others. 2025. Qwen3 technical report. Preprint, arXiv:2505.09388.
  52. Tianyi Zhang, Varsha Kishore, Felix Wu, Kilian Q. Weinberger, and Yoav Artzi. 2020. Bertscore: Evaluating text generation with bert. Preprint, arXiv:1904.09675. find
  53. Arkaitz Zubiaga, Ahmet Aker, Kalina Bontcheva, Maria Liakata, and Rob Procter. 2018. Detection and resolution of rumours in social media: A survey. ACM Comput. Surv., 51(2). find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/quaremba-2026-multilingual-cross-lingual-citation-needed/