---
title: "Mitigating Cultural Bias in LLMs via Multi-Agent Cultural Debate"
description: "To measure it, the authors build CEBiasBench, a Chinese–English benchmark of completion-style prompts covering five… 38 references."
authors: ["Qian Tan", "Lei Jiang", "Yuting Zeng", "Shuoyang Ding", "Xiaohua Xu"]
year: 2026
venue: "Findings of the Association for Computational Linguistics: ACL 2026"
source: "https://aclanthology.org/2026.findings-acl.418/"
doi: "10.18653/v1/2026.findings-acl.418"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/tan-2026-mitigating-cultural-bias-llms-multi/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/tan-2026-mitigating-cultural-bias-llms-multi/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/36/) / Mitigating Cultural Bias in LLMs via Multi-Agent Cultural Debate

# Mitigating Cultural Bias in LLMs via Multi-Agent Cultural Debate

**Authors:** Qian Tan, [Lei Jiang](https://md.research.elamur.ai/free/en/authors/lei-jiang/), Yuting Zeng, [Shuoyang Ding](https://md.research.elamur.ai/free/en/authors/shuoyang-ding/) and [Xiaohua Xu](https://md.research.elamur.ai/free/en/authors/xiaohua-xu/) · **Year:** 2026 · **Venue:** [Findings of the Association for Computational Linguistics: ACL 2026](https://md.research.elamur.ai/free/en/venues/findings/) ([Findings of the Association for Computational Linguistics 2026](https://md.research.elamur.ai/free/en/venues/findings/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.findings-acl.418/) · [PDF](https://aclanthology.org/2026.findings-acl.418.pdf) · [DOI 10.18653/v1/2026.findings-acl.418](https://doi.org/10.18653/v1/2026.findings-acl.418) · [OpenAlex](https://openalex.org/W7166843069)

**Citations:** 0 citations (OpenAlex, 2026-10-14)

## Summary
The paper asks whether prompting LLMs in Chinese instead of English reduces their Western-leaning cultural defaults, and how such bias can be reduced without retraining. To measure it, the authors build CEBiasBench, a Chinese–English benchmark of completion-style prompts covering five everyday cultural domains, such as festive food, weddings and greetings. They also propose Multi-Agent Vote (MAV), an evaluation protocol in which culturally diverse judge agents reach a majority verdict and may rule that an answer shows no bias. Their analysis shows that Chinese prompts do not remove bias. Models move toward East Asian answers, and some Western preference remains. As a remedy, they introduce Multi-Agent Cultural Debate (MACD), an inference-time method. Agents representing Western, East Asian, African, Middle Eastern and South Asian viewpoints first answer independently. They then debate over several rounds under the SCGRD strategy, which looks for shared ground while keeping meaningful differences. Finally, a summary agent merges the results. MACD raises the No Bias Rate and also transfers to the Arabic CAMeL benchmark. This suggests that explicit cultural roles help more than purely functional agent roles.

## Key points
- With a GPT-4o backbone, MACD reaches an average No Bias Rate of 57.6% (LLM-as-judge) and 86.0% (MAV). Direct generation scores 47.6% and 69.0%.
- On CEBiasBench-CN, direct generation, CoT and MAD give 38–48% East Asian-biased answers while still showing 8–14% Western bias. MACD reduces East Asian bias to about 20–21%.
- Ablations on CEBiasBench-EN (Qwen-32B evaluator) show the full system at 80.0%:
  - Removing the cultural personas lowers the score to 59.0%, and removing the consensus strategy lowers it to 61.0%.
  - Using 1, 3 and 5 agents gives 60.0%, 70.0% and 80.0%.
  - Two debate rounds work best; one round scores 53.0% and three rounds score 64.0%, which the authors attribute to over-smoothing.
- MACD answers have the highest information density (8.65 unique content words per 100 characters), compared with direct generation (8.11), CoT (6.60) and MAD (4.29).
- Limitations: coverage is restricted to five domains with discrete labels, and API costs prevented repeated runs, so the stability of the results is unverified.

*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

- [University of Science and Technology of China](https://md.research.elamur.ai/free/en/universities/university-of-science-and-technology-of-china/) — University, Hefei, China · 108 works · 112,698 citations

📚 Shelves (6): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works · [🤝 Social Sciences](https://md.research.elamur.ai/free/en/shelves/social-sciences/) · 1,772 works · [🤝 Safety Research](https://md.research.elamur.ai/free/en/shelves/social-sciences/safety-research/) · 543 works

🧭 Topics (1): [🛡️ Bias, fairness & safety](https://md.research.elamur.ai/free/en/topics/bias-fairness-safety/) · 517 works

## Cite

Cite as: Qian Tan, Lei Jiang, Yuting Zeng, Shuoyang Ding and Xiaohua Xu. 2026. Mitigating Cultural Bias in LLMs via Multi-Agent Cultural Debate. Findings of the Association for Computational Linguistics: ACL 2026.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{tan2026mitigating,
  title = {{Mitigating Cultural Bias in LLMs via Multi-Agent Cultural Debate}},
  author = {Qian Tan, Lei Jiang, Yuting Zeng, Shuoyang Ding and Xiaohua Xu},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.418},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.418}
}
```

- [← previous Mitigating Coordinate Prediction Bias from Positional Encoding Failures](https://md.research.elamur.ai/free/en/papers/tao-2026-mitigating-coordinate-prediction-bias-positional/)
- [next → Mitigating Degree Bias in Hypergraphs via Attribute-as-Structure Approach](https://md.research.elamur.ai/free/en/papers/nishide-2026-mitigating-degree-bias-hypergraphs-attribute/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/36/)

## References

38 references, 21 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Marwa Abdulhai, Gregory Serapio-Garcia, Clément Crepy, Daria Valter, John Canny, and Natasha Jaques. 2024. [Moral foundations of large language models](https://md.research.elamur.ai/free/en/papers/abdulhai-2024-moral-foundations-large-language-models/). In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 17737–17752.
2. Badr AlKhamissi, Muhammad ElNokrashy, Mai AlKhamissi, and Mona Diab. 2024. [Investigating cultural alignment of large language models](https://md.research.elamur.ai/free/en/papers/alkhamissi-2024-investigating-cultural-alignment-large-language/). arXiv preprint arXiv:2402.13231.
3. Yong Cao, Li Zhou, Seolhwa Lee, Laura Cabello, Min Chen, and Daniel Hershcovich. 2023. [Assessing cross-cultural alignment between chatgpt and human societies: An empirical study](https://md.research.elamur.ai/free/en/papers/cao-2023-assessing-cross-cultural-alignment-between/). arXiv preprint arXiv:2303.17466.
4. Souradip Chakraborty, Jiahao Qiu, Hui Yuan, Alec Koppel, Furong Huang, Dinesh Manocha, Amrit Singh Bedi, and Mengdi Wang. 2024. Maxmin-rlhf: Alignment with diverse human preferences. arXiv preprint arXiv:2402.08925. — [arXiv:2402.08925](https://arxiv.org/abs/2402.08925) [find](https://scholar.google.com/scholar?q=Souradip+Chakraborty%2C+Jiahao+Qiu%2C+Hui+Yuan%2C+Alec+Koppel%2C+Furong+Huang%2C+Dinesh+Manocha%2C+Amrit+Singh+Bedi%2C+and+Mengdi+Wang%2E+2024%2E+Maxmin%2Drlhf%3A+Alignment+with+diverse+human+preferences%2E+arXiv+preprint+ar)
5. Yu Ying Chiu, Liwei Jiang, Bill Yuchen Lin, Chan Young Park, Shuyue Stella Li, Sahithya Ravi, Mehar Bhatia, Maria Antoniak, Yulia Tsvetkov, Vered Shwartz, and 1 others. 2024. Culturalbench: a robust, diverse and challenging benchmark on measuring (the lack of) cultural knowledge of llms. [find](https://scholar.google.com/scholar?q=Yu+Ying+Chiu%2C+Liwei+Jiang%2C+Bill+Yuchen+Lin%2C+Chan+Young+Park%2C+Shuyue+Stella+Li%2C+Sahithya+Ravi%2C+Mehar+Bhatia%2C+Maria+Antoniak%2C+Yulia+Tsvetkov%2C+Vered+Shwartz%2C+and+1+others%2E+2024%2E+Culturalbench%3A+a+robust%2C+)
6. Jesse Dodge, Maarten Sap, Ana Marasović, William Agnew, Gabriel Ilharco, Dirk Groeneveld, Margaret Mitchell, and Matt Gardner. 2021. Documenting large webtext corpora: A case study on the colossal clean crawled corpus. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language P… [find](https://scholar.google.com/scholar?q=Jesse+Dodge%2C+Maarten+Sap%2C+Ana+Marasovic%CC%81%2C+William+Agnew%2C+Gabriel+Ilharco%2C+Dirk+Groeneveld%2C+Margaret+Mitchell%2C+and+Matt+Gardner%2E+2021%2E+Documenting+large+webtext+corpora%3A+A+case+study+on+the+colossal+cl)
7. Yilun Du, Shuang Li, Antonio Torralba, Joshua B Tenenbaum, and Igor Mordatch. 2023. [Improving factuality and reasoning in language models through multiagent debate](https://md.research.elamur.ai/free/en/papers/du-2023-improving-factuality-reasoning-language-models/). In Forty-first International Conference on Machine Learning.
8. Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Amy Yang, Angela Fan, and 1 others. 2024. [The llama 3 herd of models](https://md.research.elamur.ai/free/en/papers/grattafiori-2024-llama-3-herd-models/). arXiv e-prints, pages arXiv–2407.
9. Team GLM, Aohan Zeng, Bin Xu, Bowen Wang, Chenhui Zhang, Da Yin, Dan Zhang, Diego Rojas, Guanyu Feng, Hanlin Zhao, and 1 others. 2024. [Chatglm: A family of large language models from glm-130b to glm-4 all tools](https://md.research.elamur.ai/free/en/papers/glm-2024-chatglm-family-large-language-models/). arXiv preprint arXiv:2406.12793.
10. Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, and 1 others. 2025a. [Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning](https://md.research.elamur.ai/free/en/papers/deepseek-ai-2025-deepseek-r1-incentivizing-reasoning-capability/). arXiv preprint arXiv:2501.12948.
11. Geyang Guo, Tarek Naous, Hiromi Wakaki, Yukiko Nishimura, Yuki Mitsufuji, Alan Ritter, and Wei Xu. 2025b. Care: Aligning language models for regional cultural awareness. arXiv preprint arXiv:2504.05154. — [arXiv:2504.05154](https://arxiv.org/abs/2504.05154) [find](https://scholar.google.com/scholar?q=Geyang+Guo%2C+Tarek+Naous%2C+Hiromi+Wakaki%2C+Yukiko+Nishimura%2C+Yuki+Mitsufuji%2C+Alan+Ritter%2C+and+Wei+Xu%2E+2025b%2E+Care%3A+Aligning+language+models+for+regional+cultural+awareness%2E+arXiv+preprint+arXiv%3A2504%2E0515)
12. Dayeon Ki, Rachel Rudinger, Tianyi Zhou, and Marine Carpuat. 2025. [Multiple llm agents debate for equitable cultural alignment](https://md.research.elamur.ai/free/en/papers/ki-2025-multiple-llm-agents-debate-equitable/). arXiv preprint arXiv:2505.24671.
13. Cheng Li, Mengzhuo Chen, Jindong Wang, Sunayana Sitaram, and Xing Xie. 2024a. [Culturellm: Incorporating cultural differences into large language models](https://md.research.elamur.ai/free/en/papers/li-2024-culturellm-incorporating-cultural-differences-large/). Advances in Neural Information Processing Systems, 37:84799–84838.
14. Cheng Li, Damien Teney, Linyi Yang, Qingsong Wen, Xing Xie, and Jindong Wang. 2024b. Culturepark: Boosting cross-cultural understanding in large language models. Advances in Neural Information Processing Systems, 37:65183–65216. [find](https://scholar.google.com/scholar?q=Cheng+Li%2C+Damien+Teney%2C+Linyi+Yang%2C+Qingsong+Wen%2C+Xing+Xie%2C+and+Jindong+Wang%2E+2024b%2E+Culturepark%3A+Boosting+cross%2Dcultural+understanding+in+large+language+models%2E+Advances+in+Neural+Information+Process)
15. Huihan Li, Liwei Jiang, Jena D Hwang, Hyunwoo Kim, Sebastin Santy, Taylor Sorensen, Bill Yuchen Lin, Nouha Dziri, Xiang Ren, and Yejin Choi. 2024c. [Culture-gen: Revealing global cultural perception in language models through natural language prompting](https://md.research.elamur.ai/free/en/papers/li-2024-culture-gen-revealing-global-cultural/). arXiv preprint arXiv:2404.10199.
16. Aixin Liu, Bei Feng, Bing Xue, Bingxuan Wang, Bochao Wu, Chengda Lu, Chenggang Zhao, Chengqi Deng, Chenyu Zhang, Chong Ruan, and 1 others. 2024. Deepseek-v3 technical report. arXiv preprint arXiv:2412.19437. [find](https://scholar.google.com/scholar?q=Aixin+Liu%2C+Bei+Feng%2C+Bing+Xue%2C+Bingxuan+Wang%2C+Bochao+Wu%2C+Chengda+Lu%2C+Chenggang+Zhao%2C+Chengqi+Deng%2C+Chenyu+Zhang%2C+Chong+Ruan%2C+and+1+others%2E+2024%2E+Deepseek%2Dv3+technical+report%2E+arXiv+preprint+arXiv%3A2412)
17. Kelly Marchisio, Wei-Yin Ko, Alexandre Bérard, Théo Dehaze, and Sebastian Ruder. 2024. [Understanding and mitigating language confusion in llms](https://md.research.elamur.ai/free/en/papers/marchisio-2024-understanding-mitigating-language-confusion-llms/). arXiv preprint arXiv:2406.20052.
18. Reem Masoud, Ziquan Liu, Martin Ferianc, Philip C Treleaven, and Miguel Rodrigues Rodrigues. 2025. [Cultural alignment in large language models: An explanatory analysis based on hofstede’s cultural dimensions](https://md.research.elamur.ai/free/en/papers/masoud-2025-cultural-alignment-large-language-models/). In Proceedings of the 31st International Conference on Computational Linguistics, pages 847…
19. Mistral AI. 2024. Mistral large, our new flagship model. [https://mistral.ai/news/](https://mistral.ai/news/) mistral-large. Technical report. [↗](https://mistral.ai/news/)
20. Simon Münker. 2025. Cultural bias in large language models: Evaluating ai agents through moral questionnaires. arXiv preprint arXiv:2507.10073. — [arXiv:2507.10073](https://arxiv.org/abs/2507.10073) [find](https://scholar.google.com/scholar?q=Simon+M%C3%BCnker%2E+2025%2E+Cultural+bias+in+large+language+models%3A+Evaluating+ai+agents+through+moral+questionnaires%2E+arXiv+preprint+arXiv%3A2507%2E10073%2E+%E2%80%94+%5BarXiv%3A2507%2E10073%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2507%2E10073%29)
21. Rémi Munos, Michal Valko, Daniele Calandriello, Mohammad Gheshlaghi Azar, Mark Rowland, Zhaohan Daniel Guo, Yunhao Tang, Matthieu Geist, Thomas Mesnard, Côme Fiegel, and 1 others. 2024. Nash learning from human feedback. In Forty-first International Conference on Machine Learning. [find](https://scholar.google.com/scholar?q=R%C3%A9mi+Munos%2C+Michal+Valko%2C+Daniele+Calandriello%2C+Mohammad+Gheshlaghi+Azar%2C+Mark+Rowland%2C+Zhaohan+Daniel+Guo%2C+Yunhao+Tang%2C+Matthieu+Geist%2C+Thomas+Mesnard%2C+C%C3%B4me+Fiegel%2C+and+1+others%2E+2024%2E+Nash+learning+)
22. Junho Myung, Nayeon Lee, Yi Zhou, Jiho Jin, Rifki Putri, Dimosthenis Antypas, Hsuvas Borkakoty, Eunsu Kim, Carla Perez-Almendros, Abinew Ali Ayele, and 1 others. 2024. [Blend: A benchmark for llms on everyday knowledge in diverse cultures and languages](https://md.research.elamur.ai/free/en/papers/myung-2024-blend-benchmark-llms-everyday-knowledge/). Advances in Neural Information Processing Syste…
23. Tarek Naous, Michael J Ryan, Alan Ritter, and Wei Xu. 2024a. [Having beer after prayer? measuring cultural bias in large language models](https://md.research.elamur.ai/free/en/papers/naous-2023-having-beer-after-prayer-measuring/). In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 16366–16393, Bangkok, Thailand. Associati…
24. Tarek Naous, Michael J Ryan, Alan Ritter, and Wei Xu. 2024b. [Having beer after prayer? measuring cultural bias in large language models](https://md.research.elamur.ai/free/en/papers/naous-2023-having-beer-after-prayer-measuring/). In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 16366–16393.
25. Tarek Naous, Anagha Savit, Carlos Rafael Catalan, Geyang Guo, Jaehyeok Lee, Kyungdon Lee, Lheane Marie Dizon, Mengyu Ye, Neel Kothari, Sahajpreet Singh, and 1 others. 2025. Camellia: Benchmarking cultural biases in llms for asian languages. arXiv preprint arXiv:2510.05291. — [arXiv:2510.05291](https://arxiv.org/abs/2510.05291) [find](https://scholar.google.com/scholar?q=Tarek+Naous%2C+Anagha+Savit%2C+Carlos+Rafael+Catalan%2C+Geyang+Guo%2C+Jaehyeok+Lee%2C+Kyungdon+Lee%2C+Lheane+Marie+Dizon%2C+Mengyu+Ye%2C+Neel+Kothari%2C+Sahajpreet+Singh%2C+and+1+others%2E+2025%2E+Camellia%3A+Benchmarking+cult)
26. Tarek Naous and Wei Xu. 2025. [On the origin of cultural biases in language models: From pre-training data to linguistic phenomena](https://md.research.elamur.ai/free/en/papers/naous-2025-origin-cultural-biases-language-models/). In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Lon…
27. OpenAI. 2024. Gpt-4o: Our most advanced ai model. Accessed: 2024-07-20. [find](https://scholar.google.com/scholar?q=OpenAI%2E+2024%2E+Gpt%2D4o%3A+Our+most+advanced+ai+model%2E+Accessed%3A+2024%2D07%2D20%2E)
28. OpenAI. 2025. Introducing gpt-5. Technical report, OpenAI. Technical report. [find](https://scholar.google.com/scholar?q=OpenAI%2E+2025%2E+Introducing+gpt%2D5%2E+Technical+report%2C+OpenAI%2E+Technical+report%2E)
29. Siddhesh Pawar, Junyeong Park, Jiho Jin, Arnav Arora, Junho Myung, Srishti Yadav, Faiz Ghifari Haznitrama, Inhwa Song, Alice Oh, and Isabelle Augenstein. 2025. [Survey of cultural awareness in language models: Text and beyond](https://md.research.elamur.ai/free/en/papers/pawar-2025-survey-cultural-awareness-language-models/). Computational Linguistics, pages 1–96.
30. Haoyi Qiu, Alexander Richard Fabbri, Divyansh Agarwal, Kung-Hsiang Huang, Sarah Tan, Nanyun Peng, and Chien-Sheng Wu. 2025. Evaluating cultural and social awareness of llm web agents. In Findings of the Association for Computational Linguistics: NAACL 2025, pages 3978–4005. [find](https://scholar.google.com/scholar?q=Haoyi+Qiu%2C+Alexander+Richard+Fabbri%2C+Divyansh+Agarwal%2C+Kung%2DHsiang+Huang%2C+Sarah+Tan%2C+Nanyun+Peng%2C+and+Chien%2DSheng+Wu%2E+2025%2E+Evaluating+cultural+and+social+awareness+of+llm+web+agents%2E+In+Findings+of+t)
31. Shyam Sundhar Ramesh, Yifan Hu, Iason Chaimalas, Viraj Mehta, Pier Giuseppe Sessa, Haitham Bou Ammar, and Ilija Bogunovic. 2024. Group robust preference optimization in reward-free rlhf. Advances in Neural Information Processing Systems, 37:37100– 37137. [find](https://scholar.google.com/scholar?q=Shyam+Sundhar+Ramesh%2C+Yifan+Hu%2C+Iason+Chaimalas%2C+Viraj+Mehta%2C+Pier+Giuseppe+Sessa%2C+Haitham+Bou+Ammar%2C+and+Ilija+Bogunovic%2E+2024%2E+Group+robust+preference+optimization+in+reward%2Dfree+rlhf%2E+Advances+in+N)
32. Aida Ramezani and Yang Xu. 2023. [Knowledge of cultural moral norms in large language models](https://md.research.elamur.ai/free/en/papers/ramezani-2023-knowledge-cultural-moral-norms-large/). arXiv preprint arXiv:2306.01857.
33. Virgile Rennard, Christos Xypolopoulos, and Michalis Vazirgiannis. 2025. Bias in the mirror: Are llms opinions robust to their own adversarial attacks. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2128–2143. [find](https://scholar.google.com/scholar?q=Virgile+Rennard%2C+Christos+Xypolopoulos%2C+and+Michalis+Vazirgiannis%2E+2025%2E+Bias+in+the+mirror%3A+Are+llms+opinions+robust+to+their+own+adversarial+attacks%2E+In+Proceedings+of+the+63rd+Annual+Meeting+of+the)
34. Weiyan Shi, Ryan Li, Yutong Zhang, Caleb Ziems, Sunny Yu, Raya Horesh, Rogério Abreu De Paula, and Diyi Yang. 2024. [CultureBank: An online community-driven knowledge base towards culturally aware language technologies](https://md.research.elamur.ai/free/en/papers/shi-2024-culturebank-online-community-driven-knowledge/). In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 4…
35. Shivalika Singh, Angelika Romanou, Clémentine Fourrier, David Ifeoluwa Adelani, Jian Gang Ngui, Daniel Vila-Suero, Peerat Limkonchotiwat, Kelly Marchisio, Wei Qi Leong, Yosephine Susanto, and 1 others. 2025. [Global mmlu: Understanding and addressing cultural and linguistic biases in multilingual eva…](https://md.research.elamur.ai/free/en/papers/singh-2025-global-mmlu-understanding-addressing-cultural/)
36. Nicholas Sukiennik, Chen Gao, Fengli Xu, and Yong Li. 2025. An evaluation of cultural value alignment in llm. arXiv preprint arXiv:2504.08863. and Zhichao Lu. 2025. Mitigating social bias in large language models: A multi-objective approach within a multi-agent framework. In Proceedings of the AAAI… — [arXiv:2504.08863](https://arxiv.org/abs/2504.08863) [find](https://scholar.google.com/scholar?q=Nicholas+Sukiennik%2C+Chen+Gao%2C+Fengli+Xu%2C+and+Yong+Li%2E+2025%2E+An+evaluation+of+cultural+value+alignment+in+llm%2E+arXiv+preprint+arXiv%3A2504%2E08863%2E+and+Zhichao+Lu%2E+2025%2E+Mitigating+social+bias+in+large+lan)
37. An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, and 1 others. 2025. [Qwen3 technical report](https://md.research.elamur.ai/free/en/papers/yang-2025-qwen3-technical-report/). arXiv preprint arXiv:2505.09388.
38. Jing Yao, Xiaoyuan Yi, Jindong Wang, Zhicheng Dou, and Xing Xie. 2025. Caredio: Cultural alignment of llm via representativeness and distinctiveness guided data optimization. arXiv preprint arXiv:2504.08820. — [arXiv:2504.08820](https://arxiv.org/abs/2504.08820) [find](https://scholar.google.com/scholar?q=Jing+Yao%2C+Xiaoyuan+Yi%2C+Jindong+Wang%2C+Zhicheng+Dou%2C+and+Xing+Xie%2E+2025%2E+Caredio%3A+Cultural+alignment+of+llm+via+representativeness+and+distinctiveness+guided+data+optimization%2E+arXiv+preprint+arXiv%3A2504)

*Updated 1 November 2026*