Elamur Research⌕
☰
home / papers / 2026 / LLM-XTM: Enhancing Cross-Lingual Topic Models with Large Language Models

LLM-XTM: Enhancing Cross-Lingual Topic Models with Large Language Models

Paper · 2026 · Minh Chu Xuan, Tien-Phat Nguyen, Linh Ngo Van et al.

Summary

Cross-lingual topic models try to find themes shared across languages. They depend on limited bilingual resources such as seed dictionaries, document embeddings or parallel corpora, so paired topics often drift apart in meaning. Earlier LLM-assisted methods either query the model for every document, which is costly and prone to hallucination, or, as in LLM-ITL, require white-box token probabilities.

LLM-XTM is a post-hoc plug-in applied to an already trained backbone (InfoCTM, NMTM, XTRA). An LLM rewrites each topic's bilingual top-word lists several times. Following the SelfCheckGPT idea, only words that recur across these sampled refinements are kept. An MMD loss then pulls the model's topic-word distributions toward this filtered set. A second, QA-inspired loss embeds documents and refined topics with BGE-M3 and minimizes KL divergence between the model's document-topic proportions and similarity-based targets.

On English–Chinese and English–Japanese corpora (Amazon Review, ECNews, Rakuten Amazon), the authors report better coherence and cross-lingual alignment than the baselines. The method needs only black-box LLM access and relies less on dictionaries. This is useful for multilingual corpora that lack strong bilingual resources.

Key points

  • Two added objectives on top of a VAE-based backbone:
    • an MMD loss (Gaussian kernel on cosine distances) aligning topic-word distributions with LLM-refined words;
    • a KL(θ_d ∥ θ̂_d) document-level alignment loss using BGE-M3 embeddings.
  • Uncertainty is estimated by self-consistency across multiple sampled refinement paths, not by token probabilities. This makes the method usable with closed-source LLMs.
  • Sensitivity study on Amazon Review with NMTM, at refinement frequency f = 8:
    • Raising refinement rounds R from 1 to 5 lifted CNPMI from 0.0482 to 0.0562 (+16.6%). TU rose only from 0.600 to 0.627.
    • Beyond R = 5 the gains are small, so the authors recommend R ≈ 5–7.
    • More frequent LLM calls raise CNPMI but lower TU, a trade-off between coherence and topic uniqueness.
  • An LLM-as-judge evaluation (1–3 scale, mean of four runs) on XTRA showed gains in both intra-lingual coherence and cross-lingual alignment. The alignment gains were larger where the baseline was unstable.
  • Stated limitations:
    • The method can refine topics but cannot rescue a failed backbone.
    • Repeated API calls add latency.
  • Code is released at github.com/tienphat140205/LLM-XTM.

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

🎓Monash UniversityMelbourne · Australia60 works · 54,866 citations🎓Hanoi University of Science and TechnologyHanoi · Vietnam10 works · 89 citations🎓Trường Đại học Khoa học và Công nghệ Hà NộiHanoi · Vietnam8 works · 88 citations🎓VNU University of Engineering and TechnologyHanoi · Vietnam4 works · 1 citation

 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 worksGeneral Social Sciences · 227 works

🧭 Topics 2Research subjects and languages the work is about.Language models & pretraining · 3,320 worksMultilingual & cross-lingual NLP · 828 works

💡 Concepts 1Key ideas and methods explained on our concept pages.Topic modeling · 47 works

Cite

Cite as Minh Chu Xuan, Tien-Phat Nguyen, Linh Ngo Van, Dinh Viet Sang, Nguyen Thi Ngoc Diep and Trung Le. 2026. LLM-XTM: Enhancing Cross-Lingual Topic Models with Large Language Models. 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.170

@inproceedings{xuan2026llm,
  title = {{LLM-XTM: Enhancing Cross-Lingual Topic Models with Large Language Models}},
  author = {Minh Chu Xuan, Tien-Phat Nguyen, Linh Ngo Van, Dinh Viet Sang, Nguyen Thi Ngoc Diep and Trung Le},
  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.170},
  url = {https://doi.org/10.18653/v1/2026.acl-long.170}
}

References

69 references, 16 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Federico Bianchi, Silvia Terragni, and Dirk Hovy. 2021a. Pre-training is a hot topic: Contextualized document embeddings improve topic coherence. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language…
  2. Federico Bianchi, Silvia Terragni, Dirk Hovy, Debora Nozza, and Elisabetta Fersini. 2021b. Cross-lingual contextualized topic models with zero-shot learning. In Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, EACL 2021, Online…
  3. David M Blei, Andrew Y Ng, and Michael I Jordan. 2003. Latent dirichlet allocation. Journal of Machine Learning Research, pages 993–1022.
  4. Jordan L. Boyd-Graber and David M. Blei. 2012. Multilingual topic models for unaligned text. CoRR, abs/1205.2657. find
  5. Chia-Hsuan Chang, Tien-Yuan Huang, Yi-Hang Tsai, Chia-Ming Chang, and San-Yih Hwang. 2024. Refining dimensions for improving clustering-based cross-lingual topic models. CoRR, abs/2412.12433. find
  6. Chia-Hsuan Chang and San-Yih Hwang. 2021. A word embedding-based approach to cross-lingual topic modeling. Knowl. Inf. Syst., 63(6):1529–1555. find
  7. Jianlv Chen, Shitao Xiao, Peitian Zhang, Kun Luo, Defu Lian, and Zheng Liu. 2024. BGE m3-embedding: Multi-lingual, multi-functionality, multi-granularity text embeddings through self-knowledge distillation. CoRR, abs/2402.03216.
  8. Adji B Dieng, Francisco JR Ruiz, and David M Blei. 2020. Topic modeling in embedding spaces. Transactions of the Association for Computational Linguistics, pages 439–453.
  9. Tomoki Doi, Masaru Isonuma, and Hitomi Yanaka. 2024. Topic modeling for short texts with large language models. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics, ACL 2024 - Student Research Workshop, Bangkok, Thailand, August 11-16, 2024, pages 21– 33. Assoc… find
  10. Dimo Angelov. 2020. Top2vec: Distributed representations of topics. CoRR, abs/2008.09470.
  11. Damien Garreau, Wittawat Jitkrittum, and Motonobu Kanagawa. 2017. Large sample analysis of the median heuristic. arXiv preprint arXiv:1707.07269. — [arXiv:1707.07269](https://arxiv.org/abs/1707.07269) find
  12. Mikel Artetxe and Holger Schwenk. 2019. Massively multilingual sentence embeddings for zero-shot cross-lingual transfer and beyond. Trans. Assoc. Comput. Linguistics, 7:597–610.
  13. Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, and 1 others. 2024. The llama 3 herd of models. CoRR, abs/2407.21783.
  14. Arthur Gretton, Karsten M. Borgwardt, Malte J. Rasch, Bernhard Schölkopf, and Alexander J. Smola. 2012. A kernel two-sample test. J. Mach. Learn. Res., 13:723–773. find
  15. Maarten Grootendorst. 2022. Bertopic: Neural topic modeling with a class-based tf-idf procedure. Preprint, arXiv:2203.05794.
  16. Shudong Hao, Jordan L. Boyd-Graber, and Michael J. Paul. 2018. Lessons from the bible on modern topics: Low-resource multilingual topic model evaluation. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies,… find
  17. Thomas Hofmann. 1999. Probabilistic latent semantic indexing. In Proceedings of the 22nd annual international ACM SIGIR conference on Research and development in information retrieval, pages 50–57. find
  18. Alexander Miserlis Hoyle, Pranav Goel, and Philip Resnik. 2020. Improving neural topic models using knowledge distillation. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, EMNLP 2020, Online, November 16-20, 2020, pages 1752–1771. Association for Computatio… find
  19. Jagadeesh Jagarlamudi and Hal Daumé III. 2010. Extracting multilingual topics from unaligned comparable corpora. In Advances in Information Retrieval, 32nd European Conference on IR Research, ECIR 2010, Milton Keynes, UK, March 28-31, 2010. Proceedings, volume 5993 of Lecture Notes in Computer Scien… find
  20. Ziwei Ji, Nayeon Lee, Rita Frieske, Tiezheng Yu, Dan Su, Yan Xu, Etsuko Ishii, Yejin Bang, Andrea Madotto, and Pascale Fung. 2023. Survey of hallucination in natural language generation. ACM Comput. Surv., 55(12):248:1–248:38.
  21. Chi Pham Khanh, Tue Le, Anh-Cuong Le, and Linh Ngo Van. 2026a. A multi-objective surrogate framework for training neural topic models. Neurocomputing, 682:133432. find
  22. Chi Pham Khanh, Hoang Tran Vuong, Sang Dinh Viet, and Linh Ngo Van. 2026b. A framework for neural topic modeling using hierarchical clustering and contrastive learning with optimal transport. Neurocomputing, 670:132530. find
  23. Diederik P Kingma and Max Welling. 2013. Autoencoding variational bayes. In 2nd International Conference on Learning Representations. find
  24. Tue Le, Hoang Tran Vuong, Tung Nguyen, Linh Ngo Van, Dinh Viet Sang, Trung Le, and Thien Huu Nguyen. 2025. Multi-surrogate-objective optimization for neural topic models. In Findings of the Association for Computational Linguistics: EMNLP 2025, Suzhou, China, November 4-9, 2025, pages 135–151. Assoc… find
  25. Chun-Liang Li, Wei-Cheng Chang, Yu Cheng, Yiming Yang, and Barnabás Póczos. 2017. Mmd gan: Towards deeper understanding of moment matching network. In Advances in Neural Information Processing Systems (NeurIPS). find
  26. Yujia Li, Kevin Swersky, and Richard S. Zemel. 2015. Generative moment matching networks. CoRR, abs/1502.02761. find
  27. Mingsheng Long, Yue Cao, Jianmin Wang, and Michael I. Jordan. 2015. Learning transferable features with deep adaptation networks. In Proceedings of the 32nd International Conference on Machine Learning (ICML), pages 97–105. find
  28. Potsawee Manakul, Adian Liusie, and Mark J. F. Gales. 2023. Selfcheckgpt: Zero-resource black-box hallucination detection for generative large language models. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, EMNLP 2023, Singapore, December 6-10, 2023, pages… find
  29. David M. Mimno, Hanna M. Wallach, Jason Naradowsky, David A. Smith, and Andrew McCallum. 2009. Polylingual topic models. In Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing, EMNLP 2009, 6-7 August 2009, Singapore, A meeting of SIGDAT, a Special Interest Group of…
  30. Mistral AI. 2025. Mistral small 3. https://mistral.ai/news/mistral-small-3. Accessed: April 2026. ↗
  31. Yida Mu, Chun Dong, Kalina Bontcheva, and Xingyi Song. 2024. Large language models offer an alternative to the traditional approach of topic modelling. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation, LREC/COLING 2024, 20-25 M… find
  32. Krikamol Muandet, Kenji Fukumizu, Bharath K. Sriperumbudur, and Bernhard Schölkopf. 2017. Kernel mean embedding of distributions: A review and beyond. Foundations and Trends in Machine Learning, 10(1–2):1–141. find
  33. Aaron Mueller and Mark Dredze. 2021. Fine-tuning encoders for improved monolingual and zero-shot polylingual neural topic modeling. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2021, Onlin… find
  34. Feng Nan, Ran Ding, Ramesh Nallapati, and Bing Xiang. 2019. Topic modeling with wasserstein autoencoders. In Proceedings of the 57th Conference of the Association for Computational Linguistics, ACL 2019, Florence, Italy, July 28- August 2, 2019, Volume 1: Long Papers, pages 6345–6381. Association fo… find
  35. Quang Duc Nguyen, Tung Nguyen, Duc Anh Nguyen, Linh Ngo Van, Sang Dinh, and Thien Huu Nguyen. 2025a. Glocom: A short text neural topic model via global clustering context. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: H… find
  36. Thanh Vinh Nguyen, Ngo Van Dong, Minh Chu Xuan, Tung Nguyen, Linh Ngo Van, Dinh Viet Sang, and Trung Le. 2026a. Beyond coherence: Improving temporal consistency and interpretability in dynamic topic models. In Findings of the Association for Computational Linguistics: EACL, pages 3609–3629.
  37. Thong Nguyen and Anh Tuan Luu. 2021. Contrastive learning for neural topic model. In Advances in Neural Information Processing Systems, pages 11974– 11986. find
  38. Thong Thanh Nguyen, Xiaobao Wu, Xinshuai Dong, Cong-Duy T. Nguyen, See-Kiong Ng, and Anh Tuan Luu. 2024. Topic modeling as multi-objective contrastive optimization. In The Twelfth International Conference on Learning Representations, ICLR 2024, Vienna, Austria, May 7-11, 2024. OpenReview.net. find
  39. Tien Phat Nguyen, Vu Minh Ngo, Tung Nguyen, Linh Van Ngo, Duc Anh Nguyen, Sang Dinh, and Trung Le. 2025b. Xtra: Cross-lingual topic modeling with topic and representation alignments. Preprint, arXiv:2510.02788. find
  40. Truong Nguyen, Phi Van Dat, Ngan Nguyen, Linh Ngo Van, Trung Le, and Thanh Hong Nguyen. 2026b. CTPD: cross tokenizer preference distillation. In Fortieth AAAI Conference on Artificial Intelligence, Thirty-Eighth Conference on Innovative Applications of Artificial Intelligence, Sixteenth Symposium on…
  41. Tung Nguyen, Duy-Tung Doan, Huu-Thanh Luong, and Linh Ngo Van. 2025c. Sub-document neural topic model. Knowledge-Based Systems, page 114762. find
  42. Tung Nguyen, Tue Le, Hoang Tran Vuong, Quang Duc Nguyen, Duc Anh Nguyen, Linh Ngo Van, Sang Dinh, and Thien Huu Nguyen. 2025d. Sharpness-aware minimization for topic models with high-quality document representations. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the… find
  43. Tung Nguyen, Linh Ngo Van, Anh Nguyen Duc, and Sang Dinh Viet. 2026c. Global and local context in short text neural topic model. Artif. Intell., 353:104502. find
  44. Xiaochuan Ni, Jian-Tao Sun, Jian Hu, and Zheng Chen. 2009. Mining multilingual topics from wikipedia. In Proceedings of the 18th International Conference on World Wide Web, WWW 2009, Madrid, Spain, April 20-24, 2009, pages 1155–1156. ACM. find
  45. Chau Pham, Alexander Miserlis Hoyle, Simeng Sun, Philip Resnik, and Mohit Iyyer. 2024a. Topicgpt: A prompt-based topic modeling framework. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long…
  46. Duy-Tung Pham, Thien Trang Nguyen Vu, Tung Nguyen, Linh Ngo Van, Duc Anh Nguyen, and Thien Huu Nguyen. 2024b. Neuromax: Enhancing neural topic modeling via maximizing mutual information and group topic regularization. In Findings of the Association for Computational Linguistics: EMNLP 2024. find
  47. Nguyen Tien Phat, Ngo Vu Minh, Linh Ngo Van, Nguyen Thi Ngoc Diep, and Thien Huu Nguyen. 2026. Gloctm: Cross-lingual topic modeling via a global context space. In Fortieth AAAI Conference on Artificial Intelligence, Thirty-Eighth Conference on Innovative Applications of Artificial Intelligence, Sixt… find
  48. Qwen Team. 2025. Qwen3-Coder: Agentic coding in the world. https://qwenlm.github.io/blog/ qwen3-coder/. Accessed: April 2026. ↗
  49. Emil Rijcken, Floortje Scheepers, Kalliopi Zervanou, Marco Spruit, Pablo Mosteiro, and Uzay Kaymak. 2023. Towards interpreting topic models with chatgpt. In Proceedings of the 20th World Congress of the International Fuzzy Systems Association (IFSA 2023), pages 269–275, Daegu, Korea, Republic of. Pa… find
  50. Sebastian Ruder, Ivan Vulic, and Anders Søgaard. 2019. A survey of cross-lingual word embedding models. J. Artif. Intell. Res., 65:569–631. find
  51. Dino Sejdinovic, Bharath K. Sriperumbudur, Arthur Gretton, and Kenji Fukumizu. 2013. Equivalence of distance-based and rkhs-based statistics in hypothesis testing. Annals of Statistics, 41(5):2263–2291. find
  52. Bei Shi, Wai Lam, Lidong Bing, and Yinqing Xu. 2016. Detecting common discussion topics across culture from news reader comments. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016, August 7-12, 2016, Berlin, Germany, Volume 1: Long Papers. The Assoc… find
  53. Akash Srivastava and Charles Sutton. 2017. Autoencoding variational inference for topic models. In International Conference on Learning Representations. find
  54. Dominik Stammbach, Vilém Zouhar, Alexander Miserlis Hoyle, Mrinmaya Sachan, and Elliott Ash. 2023. Revisiting automated topic model evaluation with large language models. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, EMNLP 2023, Singapore, December 6-10,… find
  55. Minh-Phuc Truong, Hai An Vu, Tu Vu, Nguyen Thi Ngoc Diep, Linh Ngo Van, Thien Huu Nguyen, and Trung Le. 2025. Emo: Embedding model distillation via intra-model relation and optimal transport alignments. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages…
  56. Hai An Vu, Minh-Phuc Truong, Tu Vu, and Linh Ngo. 2026. Mol: Mixture of layers in cross-tokenizer embedding model distillation. Knowledge-Based Systems, 343:116001. find
  57. Tu Vu, Manh Do, Tung Nguyen, Ngo Van Linh, Sang Dinh, and Thien Huu Nguyen. 2025. Topic modeling for short texts via optimal transport-based clustering. In Findings of the Association for Computational Linguistics, ACL 2025, Vienna, Austria, July 27 -August 1, 2025, pages 7666–7680. Association for… find
  58. Ivan Vulic and Marie-Francine Moens. 2015. Bilingual word embeddings from non-parallel document-aligned data applied to bilingual lexicon induction. In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Langua… find
  59. Big Data, BigData 2023, Sorrento, Italy, December 15-18, 2023, pages 1236–1241. IEEE. find
  60. Xiaobao Wu, Xinshuai Dong, Thong Nguyen, Chaoqun Liu, Liangming Pan, and Anh Tuan Luu. 2023a. Infoctm: A mutual information maximization perspective of cross-lingual topic modeling. In Thirty-Seventh AAAI Conference on Artificial Intelligence, AAAI 2023, Thirty-Fifth Conference on Innovative Applica… find
  61. Xiaobao Wu, Xinshuai Dong, Thong Thanh Nguyen, and Anh Tuan Luu. 2023b. Effective neural topic modeling with embedding clustering regularization. In International Conference on Machine Learning, ICML 2023, 23-29 July 2023, Honolulu, Hawaii, USA, volume 202 of Proceedings of Machine Learning Research… find
  62. Xiaobao Wu, Chunping Li, Yan Zhu, and Yishu Miao. 2020. Learning multilingual topics with neural variational inference. In Natural Language Processing and Chinese Computing - 9th CCF International Conference, NLPCC 2020, Zhengzhou, China, October 14-18, 2020, Proceedings, Part I, volume 12430 of Lec… find
  63. Hongliang Yan, Yukang Ding, Peihua Li, Qilong Wang, Yong Xu, and Wangmeng Zuo. 2017. Mind the class weight bias: Weighted maximum mean discrepancy for unsupervised domain adaptation. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 2272–2281. find
  64. An Yang, Anfeng Li, Baosong Yang, and 1 others. 2025a. Qwen3 technical report. CoRR, abs/2505.09388.
  65. Weiwei Yang, Jordan L. Boyd-Graber, and Philip Resnik. 2019. A multilingual topic model for learning weighted topic links across corpora with low comparability. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Nat… find
  66. Hoang Tran Vuong, Tue Le, Tu Vu, Tung Nguyen, Linh Ngo Van, Sang Dinh, and Thien Huu Nguyen. 2025. Hicot: Improving neural topic models via optimal transport and contrastive learning. In Findings of the Association for Computational Linguistics, ACL 2025, Vienna, Austria, July 27 - August 1, 2025, p… find
  67. Xiaohao Yang, He Zhao, Weijie Xu, Yuanyuan Qi, Jue-qing Lu, Dinh Phung, and Lan Du. 2025b. Neural topic modeling with large language models in the loop. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2025, Vienna, Austria, July… find
  68. Han Wang, Nirmalendu Prakash, Nguyen-Khoi Hoang, Ming Shan Hee, Usman Naseem, and Roy Ka-Wei Lee. 2023. Prompting large language models for topic modeling. In IEEE International Conference on Michelle Yuan, Benjamin Van Durme, and Jordan L. Ying. 2018. Multilingual anchoring: Interactive topic model… find
  69. He Zhao, Dinh Phung, Viet Huynh, Trung Le, and Wray L. Buntine. 2021. Neural topic model via optimal transport. In 9th International Conference on Learning Representations, ICLR 2021, Virtual Event, Austria, May 3-7, 2021. OpenReview.net. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/xuan-2026-llm-xtm-enhancing-cross-lingual/