---
title: "Saudi-Alignment Benchmark — paper by Alhassoun et al. (2025)"
description: "LLMs used in a particular country must respect local customs and know basic local facts. 66 references."
authors: ["Manal Alhassoun", "Imaan Mohammed Alkhanen", "Nouf Alshalawi", "Ibtehal Baazeem", "Waleed Alsanie"]
year: 2025
venue: "Proceedings of The Third Arabic Natural Language Processing Conference"
source: "https://aclanthology.org/2025.arabicnlp-main.11/"
doi: "10.18653/v1/2025.arabicnlp-main.11"
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/alhassoun-2025-saudi-alignment-benchmark-assessing-llms/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/alhassoun-2025-saudi-alignment-benchmark-assessing-llms/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2025](https://md.research.elamur.ai/free/en/papers/2025/page/18/) / Saudi-Alignment Benchmark: Assessing LLMs Alignment with Cultural Norms and Domain Knowledge in the Saudi Context

# Saudi-Alignment Benchmark: Assessing LLMs Alignment with Cultural Norms and Domain Knowledge in the Saudi Context

**Authors:** [Manal Alhassoun](https://md.research.elamur.ai/free/en/authors/manal-alhassoun/), [Imaan Mohammed Alkhanen](https://md.research.elamur.ai/free/en/authors/imaan-mohammed-alkhanen/), [Nouf Alshalawi](https://md.research.elamur.ai/free/en/authors/nouf-alshalawi/), [Ibtehal Baazeem](https://md.research.elamur.ai/free/en/authors/ibtehal-baazeem/) and Waleed Alsanie · **Year:** 2025 · **Venue:** [Proceedings of The Third Arabic Natural Language Processing Conference](https://md.research.elamur.ai/free/en/venues/arabicnlp/) ([ArabicNLP 2025](https://md.research.elamur.ai/free/en/venues/arabicnlp/2025/)) · **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/2025.arabicnlp-main.11/) · [PDF](https://aclanthology.org/2025.arabicnlp-main.11.pdf) · [DOI 10.18653/v1/2025.arabicnlp-main.11](https://doi.org/10.18653/v1/2025.arabicnlp-main.11) · [OpenAlex](https://openalex.org/W4416037707)

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

## Summary
LLMs used in a particular country must respect local customs and know basic local facts. Arabic and Saudi evaluation resources mostly cover dialects or lifestyle trivia. The one Saudi-specific benchmark, SaudiCulture, is English-only, draws on a single source and is scored automatically.

The authors built the Saudi-Alignment Benchmark, 874 hand-written Arabic questions taken from official material such as government policies, regulations and school curricula. It has two dimensions:
- **Saudi Cultural and Ethical Norms**, covering workplace, visitors, social life and vulnerable groups.
- **Saudi Domain Knowledge**, covering history, geography, Vision 2030, and attire and cuisine.

Questions come in three formats: Fill-in-the-Blank, single-answer MCQ and Acceptable/Unacceptable judgments. Six multilingual and Arabic LLMs were tested, with some manual scoring.

GPT-4 scored highest, closely followed by ALLaM-7B and Llama-3.3-70B, while Jais-30B scored very low. Models did much better when picking from given options than when generating answers. This suggests that benchmarks built only on recognition tasks can make cultural competence look stronger than it is.

## Key points
- The benchmark has 874 Arabic items: 493 on norms and ethics and 381 on domain knowledge, roughly twice the size of SaudiCulture (441 items, only 86 at national level).
- Overall accuracy:
  - GPT-4: 83.3%
  - ALLaM-7B: 81.8%
  - Llama-3.3-70B: 81.6%
  - Jais-30B: 21.9%
- AceGPT-13B and GPT-3.5-turbo were the other two models evaluated. In one format breakdown Jais-30B reached only 0.8%.
- Multilingual models were stronger on norms than on domain knowledge. ALLaM-7B led on domain knowledge, which the authors link to its regionally targeted alignment.
- The authors partly attribute Jais-30B's weakness to its pretraining data being only about 29% Arabic. They conclude that model quality and training matter more than whether a model is "Arabic-focused".

## Interesting ideas
- Testing the same content with both option-picking and free-generation formats shows when a model recognizes an answer it cannot produce on its own, so recognition scores alone are not trusted.
- Normative content was taken only from official, citable references rather than annotator opinion, which reduces subjectivity in judgments that are otherwise hard to define.
- Scenario items asking whether a described behaviour is acceptable or unacceptable turn cultural knowledge into a binary judgment over a situation.
- Manual checking was added because automatic string matching can mark correctly but differently worded answers as wrong.
- The authors plan to use LLMs to scale question generation and grading, with safeguards so a model never judges its own output.

## For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

- **I1** (answer): School-test and fable-QA readers could be scored on paired MCQ and open-answer versions of each item, exposing cases where elimination among options masks missing understanding.
- **I2** (world): Compiled knowledge about values and social norms could carry a source pointer for every entry, keeping norm-based inferences explainable and auditable.
- **I3** (frames): Short scenarios could be parsed into who-did-what frames and checked against norm rules, giving a test set for whether the pragmatics layer can evaluate actions, not just extract them.
- **I4** (learn): Evaluation could compare generated answers at the meaning-graph level, with human spot checks, instead of by surface overlap, avoiding false negatives on paraphrases.
- **I5** (learn): In LLM teacher tiers, the model that produces silver labels or questions should be separated from the one that grades them, acting as a control against self-confirming errors.

*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 Abdulaziz City for Science and Technology](https://md.research.elamur.ai/free/en/labs/king-abdulaziz-city-for-science-and-technology/) — Research institute, Riyadh, Saudi Arabia · 8 works · 74 citations

📚 Shelves (6): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 17,450 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 17,296 works · [🤝 Social Sciences](https://md.research.elamur.ai/free/en/shelves/social-sciences/) · 1,954 works · [🤝 Safety Research](https://md.research.elamur.ai/free/en/shelves/social-sciences/safety-research/) · 626 works

🧭 Topics (3): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,894 works · [Semitic languages](https://md.research.elamur.ai/free/en/topics/semitic-languages/) · 3,066 works · [📏 Evaluation & benchmarks](https://md.research.elamur.ai/free/en/topics/evaluation-benchmarks/) · 1,341 works

## Cite

Cite as: Manal Alhassoun, Imaan Mohammed Alkhanen, Nouf Alshalawi, Ibtehal Baazeem and Waleed Alsanie. 2025. Saudi-Alignment Benchmark: Assessing LLMs Alignment with Cultural Norms and Domain Knowledge in the Saudi Context. Proceedings of The Third Arabic Natural Language Processing Conference.

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

```bibtex
@inproceedings{alhassoun2025saudi,
  title = {{Saudi-Alignment Benchmark: Assessing LLMs Alignment with Cultural Norms and Domain Knowledge in the Saudi Context}},
  author = {Manal Alhassoun, Imaan Mohammed Alkhanen, Nouf Alshalawi, Ibtehal Baazeem and Waleed Alsanie},
  booktitle = {Proceedings of The Third Arabic Natural Language Processing Conference},
  year = {2025},
  doi = {10.18653/v1/2025.arabicnlp-main.11},
  url = {https://doi.org/10.18653/v1/2025.arabicnlp-main.11}
}
```

- [← previous Sarc7: Evaluating Sarcasm Detection and Generation with Seven Types and Emotion-Informed Techniques](https://md.research.elamur.ai/free/en/papers/xiong-2025-sarc7-evaluating-sarcasm-detection-generation/)
- [next → Sawyer, Daniel. 2024. How to Read Middle English Poetry. Oxford: Oxford University Press. Pp. 240. ISBN 9780198895237.](https://md.research.elamur.ai/free/en/papers/atkinson-2025-sawyer-daniel-2024-read-middle/)
- [↑ back to papers of 2025](https://md.research.elamur.ai/free/en/papers/2025/page/18/)

## References

66 references, 24 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Shahad Al-Khalifa and Hend Al-Khalifa. 2024. [The qiyas benchmark: Measuring ChatGPT mathematical and language understanding in Arabic](https://md.research.elamur.ai/free/en/papers/khalifa-2024-qiyas-benchmark-measuring-chatgpt-mathematical/). In Proceedings of the 7th International Conference on Natural Language and Speech Processing (ICNLSP 2024), pages 343–351.
2. Emad A. Alghamdi, Reem Masoud, Deema Alnuhait, Afnan Y. Alomairi, Ahmed Ashraf, and Mohamed Zaytoon. 2025. [AraTrust: An evaluation of trustworthiness for LLMs in Arabic](https://md.research.elamur.ai/free/en/papers/alghamdi-2025-aratrust-evaluation-trustworthiness-llms-arabic/). In Proceedings of the 31st International Conference on Computational Linguistics, pages 8664–8679.
3. 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/). In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 12404–12422.
4. Ebtesam Almazrouei, Ruxandra Cojocaru, Michele Baldo, Quentin Malartic, Hamza Alobeidli, Daniele Mazzotta, Guilherme Penedo, Giulia Campesan, Mugariya Farooq, Maitha Alhammadi, and 1 others. 2023. [Alghafa evaluation benchmark for arabic language models](https://md.research.elamur.ai/free/en/papers/almazrouei-2023-alghafa-evaluation-benchmark-arabic-language/). In Proceedings of ArabicNLP 2023, pages 244–275.
5. While this benchmark aims to enhance the assessment of LLMs for Saudi cultural alignment, we Shane Arora, Marzena Karpinska, Hung-Ting Chen, Ipsita Bhattacharjee, Mohit Iyyer, and Eunsol Choi. 2025. [CaLMQA: Exploring culturally specific long-form question answering across 23 languages](https://md.research.elamur.ai/free/en/papers/arora-2025-calmqa-exploring-culturally-specific-long/). In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 11772–11817, Vienna, Austria. Association for Computational Linguistics.
6. Bureau of Experts of Council of Ministers. b. Disability rights law (royal decree no. m/27, 11/2/1445ah). [https://laws.boe.gov.sa/BoeLaws/Laws/LawDetails/](https://laws.boe.gov.sa/BoeLaws/Laws/LawDetails/) e52b691a-785c-42a7-8916-b07d00e4fd38/1. Accessed: 2025-05-04. [↗](https://laws.boe.gov.sa/BoeLaws/Laws/LawDetails/)
7. Yasser Ashraf, Yuxia Wang, Bin Gu, Preslav Nakov, and Timothy Baldwin. 2025. [Arabic dataset for llm safeguard evaluation](https://md.research.elamur.ai/free/en/papers/ashraf-2025-arabic-dataset-llm-safeguard-evaluation/). In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), pages 5529–5546.
8. Bureau of Experts of Council of Ministers. c. Elderly rights and care law (royal decree no. m/47, 3/6/1443ah). [https://laws.boe.gov.sa/BoeLaws/Laws/LawDetails/](https://laws.boe.gov.sa/BoeLaws/Laws/LawDetails/) 3c63e654-4046-468d-93fd-ae1a00de13be/1. Accessed: 2025-05-04. [↗](https://laws.boe.gov.sa/BoeLaws/Laws/LawDetails/)
9. Authority of People with Disability. The implementing regulations of the rights of persons with disabilities law. [https://apd.gov.sa/web/content/38080](https://apd.gov.sa/web/content/38080). Accessed: 2025-05-04. [↗](https://apd.gov.sa/web/content/38080)
10. Lama Ayash, Hassan Alhuzali, Ashwag Alasmari, and Sultan Aloufi. 2025. Saudiculture: A benchmark for evaluating large language models’ cultural competence within saudi arabia. Journal of King Saud University Computer and Information Sciences, 37(6):123. [find](https://scholar.google.com/scholar?q=Lama+Ayash%2C+Hassan+Alhuzali%2C+Ashwag+Alasmari%2C+and+Sultan+Aloufi%2E+2025%2E+Saudiculture%3A+A+benchmark+for+evaluating+large+language+models%E2%80%99+cultural+competence+within+saudi+arabia%2E+Journal+of+King+Saud+Uni)
11. M Saiful Bari, Yazeed Alnumay, Norah A. Alzahrani, Nouf M. Alotaibi, Hisham A. Alyahya, Sultan Al-Rashed, Faisal A. Mirza, Shaykhah Z. Alsubaie, Hassan A. Alahmed, Ghadah Alabduljabbar, Raghad Alkhathran, Yousef Almushayqih, Raneem Alnajim, Salman Alsubaihi, Maryam Al Mansour, Majed Alrubaian, Ali Alammari, Zaki Alawami, Abdulmohsen Al-Thubaity, and 6 others. 2024. [Allam: Large language models for arabic and english](https://md.research.elamur.ai/free/en/papers/bari-2024-allam-large-language-models-arabic/). Preprint, arXiv:2407.15390. [↗](https://arxiv.org/abs/2407.15390)
12. Bureau of Experts at the Council of Ministers. a. Labor law. [https://laws.boe.gov.sa/BoeLaws/Laws/LawDetails/](https://laws.boe.gov.sa/BoeLaws/Laws/LawDetails/) 08381293-6388-48e2-8ad2-a9a700f2aa94/1. Accessed: 2025-03-15; Royal Decree No. M/51 dated 23/08/1426 AH. [↗](https://laws.boe.gov.sa/BoeLaws/Laws/LawDetails/)
13. Bureau of Experts at the Council of Ministers. b. National policy for the promotion of equal opportunity and treatment in employment and occupation (royal decree no. 416, 17/06/1444ah). [https://uqn.gov.sa/?p=21527](https://uqn.gov.sa/?p=21527). Accessed: 2025-03-15. [↗](https://uqn.gov.sa/?p=21527)
14. Bureau of Experts at the Council of Ministers. c. Public decency regulations (royal decree no. 444, 04/08/1440ah). [https://laws.boe.gov.sa/BoeLaws/Laws/LawDetails/](https://laws.boe.gov.sa/BoeLaws/Laws/LawDetails/) e52b691a-785c-42a7-8916-b07d00e4fd38/1. Accessed: 2025-04-29. [↗](https://laws.boe.gov.sa/BoeLaws/Laws/LawDetails/)
15. Bureau of Experts of Council of Ministers. a. Child protection law (royal decree no. m/14, 3/2/1436ah). [https://laws.boe.gov.sa/BoeLaws/Laws/LawDetails/](https://laws.boe.gov.sa/BoeLaws/Laws/LawDetails/) 2e1544fa-0dfb-43bb-b0a7-a0c100f9496d/1. Accessed: 2025-05-04. [↗](https://laws.boe.gov.sa/BoeLaws/Laws/LawDetails/)
16. Chen-Chi Chang, Ching-Yuan Chen, Hung-Shin Lee, and Chih-Cheng Lee. 2024. Benchmarking cognitive domains for llms: Insights from taiwanese hakka culture. In 2024 27th Conference of the Oriental COCOSDA International Committee for the Coordination and Standardisation of Speech Databases and Assessment Techniques (O-COCOSDA), pages 1–6. [find](https://scholar.google.com/scholar?q=Chen%2DChi+Chang%2C+Ching%2DYuan+Chen%2C+Hung%2DShin+Lee%2C+and+Chih%2DCheng+Lee%2E+2024%2E+Benchmarking+cognitive+domains+for+llms%3A+Insights+from+taiwanese+hakka+culture%2E+In+2024+27th+Conference+of+the+Oriental+COCOSD)
17. Anastasiia Demidova, Hanin Atwany, Nour Rabih, Sanad Sha’ban, and Muhammad Abdul-Mageed. 2024. [John vs. ahmed: Debate-induced bias in multilingual LLMs](https://md.research.elamur.ai/free/en/papers/demidova-2024-john-vs-ahmed-debate-induced/). In Proceedings of the Second Arabic Natural Language Processing Conference, pages 193–209, Bangkok, Thailand. Association for Computational Linguistics.
18. Ashutosh Dwivedi, Pradhyumna Lavania, and Ashutosh Modi. 2023. Eticor: Corpus for analyzing llms for etiquettes. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 6921–6931. [find](https://scholar.google.com/scholar?q=Ashutosh+Dwivedi%2C+Pradhyumna+Lavania%2C+and+Ashutosh+Modi%2E+2023%2E+Eticor%3A+Corpus+for+analyzing+llms+for+etiquettes%2E+In+Proceedings+of+the+2023+Conference+on+Empirical+Methods+in+Natural+Language+Processi)
19. Faris Hijazi, Somayah Alharbi, Abdulaziz AlHussein, Harethah Shairah, Reem Alzahrani, Hebah Al-shamlan, George Turkiyyah, and Omar Knio. 2024. [Arablegaleval: A multitask benchmark for assessing arabic legal knowledge in large language models](https://md.research.elamur.ai/free/en/papers/hijazi-2024-arablegaleval-multitask-benchmark-assessing-arabic/). In Proceedings of The Second Arabic Natural Language Processing Conference, pages 225–249.
20. Huang Huang, Fei Yu, Jianqing Zhu, Xuening Sun, Hao Cheng, Song Dingjie, Zhihong Chen, Mosen Alharthi, Bang An, Juncai He, Ziche Liu, Junying Chen, Jianquan Li, Benyou Wang, Lian Zhang, Ruoyu Sun, Xiang Wan, Haizhou Li, and Jinchao Xu. 2024. [AceGPT, localizing large language models in Arabic](https://md.research.elamur.ai/free/en/papers/huang-2024-acegpt-localizing-large-language-models/). In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), pages 8139–8163, Mexico City, Mexico. Association for Computational Linguistics.
21. Human Rights Commission. 2023. Human rights in saudi arabia. [https://www.hrc.gov.sa/website/](https://www.hrc.gov.sa/website/) hrc-in-ksa. Accessed: 2025-05-11. [↗](https://www.hrc.gov.sa/website/)
22. Amr Keleg. 2025. [Llm alignment for the arabs: A homogenous culture or diverse ones](https://md.research.elamur.ai/free/en/papers/keleg-2025-llm-alignment-arabs-homogenous-culture/). In Proceedings of the 3rd Workshop on Cross-Cultural Considerations in NLP (C3NLP 2025), pages 1–9.
23. Amr Keleg and Walid Magdy. 2023. Dlama: A framework for curating culturally diverse facts for probing the knowledge of pretrained language models. In Findings of the Association for Computational Linguistics: ACL 2023, pages 6245–6266. [find](https://scholar.google.com/scholar?q=Amr+Keleg+and+Walid+Magdy%2E+2023%2E+Dlama%3A+A+framework+for+curating+culturally+diverse+facts+for+probing+the+knowledge+of+pretrained+language+models%2E+In+Findings+of+the+Association+for+Computational+Ling)
24. King Abdulaziz Foundation and Saudi Ministry of Culture. 2022. Saudi fashion guideline. [https://www.foundingday.sa/assets/](https://www.foundingday.sa/assets/) dlyl-alazya-aarby.pdf. Accessed: 2025-05-01. [↗](https://www.foundingday.sa/assets/)
25. King Abdulaziz Foundation and Saudi Ministry of Culture. 2023. Saudi culinary guideline. [https://www.foundingday.sa/assets/](https://www.foundingday.sa/assets/) foundingday-culinary-guideline.pdf. Accessed: 2025-05-01. [↗](https://www.foundingday.sa/assets/)
26. King Abdulaziz Public Library. Encyclopedia of the Kingdom of Saudi Arabia. [https://saudiency.kapl.org.sa](https://saudiency.kapl.org.sa). Accessed: 2025-04-28. [↗](https://saudiency.kapl.org.sa)
27. Mohamed Bayan Kmainasi, Rakif Khan, Ali Ezzat Shahroor, Boushra Bendou, Maram Hasanain, and Firoj Alam. 2024. Native vs non-native language prompting: A comparative analysis. In International Conference on Web Information Systems Engineering, pages 406–420. Springer. [find](https://scholar.google.com/scholar?q=Mohamed+Bayan+Kmainasi%2C+Rakif+Khan%2C+Ali+Ezzat+Shahroor%2C+Boushra+Bendou%2C+Maram+Hasanain%2C+and+Firoj+Alam%2E+2024%2E+Native+vs+non%2Dnative+language+prompting%3A+A+comparative+analysis%2E+In+International+Conferen)
28. Fajri Koto, Haonan Li, Sara Shatnawi, Jad Doughman, Abdelrahman Sadallah, Aisha Alraeesi, Khalid Almubarak, Zaid Alyafeai, Neha Sengupta, Shady Shehata, Nizar Habash, Preslav Nakov, and Timothy Baldwin. 2024. [ArabicMMLU: Assessing massive multitask language understanding in Arabic](https://md.research.elamur.ai/free/en/papers/koto-2024-arabicmmlu-assessing-massive-multitask-language/). In Findings of the Association for Computational Linguistics: ACL 2024, pages 5622–5640, Bangkok, Thailand. Association for Computational Linguistics.
29. Jiyoung Lee, Minwoo Kim, Seungho Kim, Junghwan Kim, Seunghyun Won, Hwaran Lee, and Edward Choi. 2024. KorNAT: LLM alignment benchmark for Korean social values and common knowledge. In Findings of the Association for Computational Linguistics: ACL 2024, pages 11177–11213, Bangkok, Thailand. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Jiyoung+Lee%2C+Minwoo+Kim%2C+Seungho+Kim%2C+Junghwan+Kim%2C+Seunghyun+Won%2C+Hwaran+Lee%2C+and+Edward+Choi%2E+2024%2E+KorNAT%3A+LLM+alignment+benchmark+for+Korean+social+values+and+common+knowledge%2E+In+Findings+of+the+)
30. Chen Liu, Fajri Koto, Timothy Baldwin, and Iryna Gurevych. 2024. [Are multilingual llms culturally-diverse reasoners? an investigation into multicultural proverbs and sayings](https://md.research.elamur.ai/free/en/papers/liu-2023-multilingual-llms-culturally-diverse-reasoners/). In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), pages 2016–2039.
31. Meta AI. 2024. Introducing llama 3: The next generation of open foundation models. [https://ai.meta.com/blog/meta-llama-3/](https://ai.meta.com/blog/meta-llama-3/). Accessed: 2025-06-16. [↗](https://ai.meta.com/blog/meta-llama-3/)
32. Ministry of Education. 2024a. Applications In Law: Third Year of High School. Saudi Ministry of Education. Accessed: 2025-04-28. [find](https://scholar.google.com/scholar?q=Ministry+of+Education%2E+2024a%2E+Applications+In+Law%3A+Third+Year+of+High+School%2E+Saudi+Ministry+of+Education%2E+Accessed%3A+2025%2D04%2D28%2E)
33. Ministry of Education. 2024b. Life Skills: Third Year of High School. Saudi Ministry of Education. Accessed: 2025-04-28. [find](https://scholar.google.com/scholar?q=Ministry+of+Education%2E+2024b%2E+Life+Skills%3A+Third+Year+of+High+School%2E+Saudi+Ministry+of+Education%2E+Accessed%3A+2025%2D04%2D28%2E)
34. Ministry of Education. 2024c. Social Studies: Grade 5, Second Semester. Saudi Ministry of Education. Accessed: 2025-05-05. [find](https://scholar.google.com/scholar?q=Ministry+of+Education%2E+2024c%2E+Social+Studies%3A+Grade+5%2C+Second+Semester%2E+Saudi+Ministry+of+Education%2E+Accessed%3A+2025%2D05%2D05%2E)
35. Ministry of Education. 2024d. Social Studies: Grade 6. Saudi Ministry of Education. Accessed: 2025-05-07. [find](https://scholar.google.com/scholar?q=Ministry+of+Education%2E+2024d%2E+Social+Studies%3A+Grade+6%2E+Saudi+Ministry+of+Education%2E+Accessed%3A+2025%2D05%2D07%2E)
36. Ministry of Education. 2024e. Social Studies: Second Year of High School. Saudi Ministry of Education. Accessed: 2025-05-07. [find](https://scholar.google.com/scholar?q=Ministry+of+Education%2E+2024e%2E+Social+Studies%3A+Second+Year+of+High+School%2E+Saudi+Ministry+of+Education%2E+Accessed%3A+2025%2D05%2D07%2E)
37. Ministry of Education. 2024f. Social Studies: Third Year of Middle School. Saudi Ministry of Education. Accessed: 2025-05-05. [find](https://scholar.google.com/scholar?q=Ministry+of+Education%2E+2024f%2E+Social+Studies%3A+Third+Year+of+Middle+School%2E+Saudi+Ministry+of+Education%2E+Accessed%3A+2025%2D05%2D05%2E)
38. Ministry of Foreign Affairs. Ksa history. [https://www.mofa.gov.sa/en/ksa/Pages/history.aspx](https://www.mofa.gov.sa/en/ksa/Pages/history.aspx). Accessed: 2025-05-07. [↗](https://www.mofa.gov.sa/en/ksa/Pages/history.aspx)
39. Ministry of Health. 2024. Dress code regulations. [https://www.moh.gov.sa/Documents/](https://www.moh.gov.sa/Documents/) rules-MOH-Client.pdf. Accessed: 2025-03-15. [↗](https://www.moh.gov.sa/Documents/)
40. Ministry of Human Resources and Social Development. a. Implementing regulation of the child protection law issued under ministerial resolution no. (182054) dated 09/10/1443 ah. [https://www.hrsd.gov.sa/sites/default/](https://www.hrsd.gov.sa/sites/default/) files/2023-02/30012023_repaired_0.pdf. Accessed: 2025-05-04. [↗](https://www.hrsd.gov.sa/sites/default/)
41. Ministry of Human Resources and Social Development. b. Implementing regulation of the child protection law issued under ministerial resolution no. (56386) dated 16/06/1436 ah. [http://bit.ly/45ILbuo](http://bit.ly/45ILbuo). Accessed: 2025-05-04. [↗](http://bit.ly/45ILbuo)
42. Ministry of Human Resources and Social Development. 2021. Guidance manual for the code of work ethics. [https://www.hrsd.gov.sa/knowledge-centre/](https://www.hrsd.gov.sa/knowledge-centre/) decisions-and-regulations/ regulation-and-procedures/838883. Accessed: 2025-03-15. [↗](https://www.hrsd.gov.sa/knowledge-centre/)
43. Ministry of Human Resources and Social Development. 2025. Regulations for announcing job vacancies and conducting job interviews. [https://bit.ly/](https://bit.ly/) 4eohOzB. Ministerial Resolution, Kingdom of Saudi Arabia. [↗](https://bit.ly/)
44. Basel Mousi, Nadir Durrani, Fatema Ahmad, Md Arid Hasan, Maram Hasanain, Tameem Kabbani, Fahim Dalvi, Shammur Absar Chowdhury, and Firoj Alam. 2025. [Aradice: Benchmarks for dialectal and cultural capabilities in llms](https://md.research.elamur.ai/free/en/papers/mousi-2025-aradice-benchmarks-dialectal-cultural-capabilities/). In Proceedings of the 31st International Conference on Computational Linguistics, pages 4186–4218.
45. Ahmad Mustapha, Hadi Al-Khansa, Hadi Al-Mubasher, Aya Mourad, Ranam Hamoud, Hasan El-Husseini, Marwah Al-Sakkaf, and Mariette Awad. 2024. [Arastem: A native arabic multiple choice question benchmark for evaluating llms knowledge in stem subjects](https://md.research.elamur.ai/free/en/papers/mustapha-2024-arastem-native-arabic-multiple-choice/). arXiv preprint arXiv:2501.00559. [↗](https://arxiv.org/abs/2501.00559)
46. 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 Systems, 37:78104–78146.
47. El Moatez Billah Nagoudi, AbdelRahim Elmadany, Ahmed Oumar El-Shangiti, and Muhammad Abdul-Mageed. 2023. [Dolphin: A challenging and diverse benchmark for Arabic NLG](https://md.research.elamur.ai/free/en/papers/nagoudi-2023-dolphin-challenging-diverse-benchmark-arabic/). In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 1404–1422, Singapore. Association for Computational Linguistics.
48. Tarek Naous, Michael J Ryan, Alan Ritter, and Wei Xu. 2024. [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.
49. National Center for Vegetation Development and Combating Desertification. About salma geop-ark. [https://ksasalmageopark.ncvc.gov.sa/](https://ksasalmageopark.ncvc.gov.sa/) ar/about.html/. Accessed: 2025-05-05. [↗](https://ksasalmageopark.ncvc.gov.sa/)
50. OpenAI, Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, Red Avila, Igor Babuschkin, Suchir Balaji, Valerie Balcom, Paul Baltescu, Haiming Bao, Mohammad Bavarian, Jeff Belgum, and 262 others. 2024. [Gpt-4 technical report](https://md.research.elamur.ai/free/en/papers/openai-2023-gpt-4-technical-report/). Preprint, arXiv:2303.08774. [↗](https://arxiv.org/abs/2303.08774)
51. Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, John Schulman, Jacob Hilton, Fraser Kelton, Luke Miller, Maddie Simens, Amanda Askell, Peter Welinder, Paul F Christiano, Jan Leike, and Ryan Lowe. 2022. [Training language models to follow instructions with human feedback](https://md.research.elamur.ai/free/en/papers/ouyang-2022-training-language-models-follow-instructions/). In Advances in Neural Information Processing Systems, volume 35, pages 27730–27744. Curran Associates, Inc. [↗](https://doi.org/10.52202/068431-2011)
52. 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, 51(3):907–1004.
53. Zhaozhi Qian, Faroq Altam, Muhammad Alqurishi, and Riad Souissi. 2024. [Cameleval: Advancing culturally aligned arabic language models and benchmarks](https://md.research.elamur.ai/free/en/papers/qian-2024-cameleval-advancing-culturally-aligned-arabic/). arXiv preprint arXiv:2409.12623. [↗](https://arxiv.org/abs/2409.12623)
54. 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/). In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 428–446, Toronto, Canada. Association for Computational Linguistics.
55. Nicholas Sadjoli, Tim Siefken, Atin Ghosh, Yifan Mai, and Daniel Dahlmeier. 2025. Optimization before evaluation: Evaluation with unoptimized prompts can be misleading. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track), pages 619–638. [find](https://scholar.google.com/scholar?q=Nicholas+Sadjoli%2C+Tim+Siefken%2C+Atin+Ghosh%2C+Yifan+Mai%2C+and+Daniel+Dahlmeier%2E+2025%2E+Optimization+before+evaluation%3A+Evaluation+with+unoptimized+prompts+can+be+misleading%2E+In+Proceedings+of+the+63rd+Annu)
56. Saudi Human Rights Commission. Equal opportunities. [https://www.hrc.gov.sa/website/](https://www.hrc.gov.sa/website/) hrc-in-ksa/1. Accessed: 2025-03-15. [↗](https://www.hrc.gov.sa/website/)
57. Saudi National Platform. 2025a. Elderly. [https://my.gov.sa/en/content/elderly](https://my.gov.sa/en/content/elderly). Accessed: 2025-05-01. [↗](https://my.gov.sa/en/content/elderly)
58. Saudi National Platform. 2025b. Rights of people with disabilities. [https://my.gov.sa/en/](https://my.gov.sa/en/) content/disabilities. Accessed: 2025-05-01. [↗](https://my.gov.sa/en/)
59. Saudi National Platform. 2025c. Women empowerment. [https://my.gov.sa/en/content/](https://my.gov.sa/en/content/) women-empowering. Accessed: 2025-05-01. [↗](https://my.gov.sa/en/content/)
60. Saudi Tourism Authority. Visit saudi. [https://www.visitsaudi.com/](https://www.visitsaudi.com/). Accessed: 2025-05-05. [↗](https://www.visitsaudi.com/)
61. Saudi Tourism Authority. 2025. Saudi culture and customs. [https://www.visitsaudi.com/](https://www.visitsaudi.com/) ar/stories/saudi-culture-and-customs. Accessed 18 June 2025. [↗](https://www.visitsaudi.com/)
62. Saudi Vision 2030. 2025a. Saudi vision 2030 (official document). [https://www.vision2030.gov.sa/](https://www.vision2030.gov.sa/) media/5ptbkbxn/saudi_vision2030_ar.pdf. Accessed: 2025-05-08. [↗](https://www.vision2030.gov.sa/)
63. Saudi Vision 2030. 2025b. Vision 2030 – overview. [https://www.vision2030.gov.sa/](https://www.vision2030.gov.sa/). Accessed: 2025-05-08. [↗](https://www.vision2030.gov.sa/)
64. Neha Sengupta, Sunil Kumar Sahu, Bokang Jia, Satheesh Katipomu, Haonan Li, Fajri Koto, William Marshall, Gurpreet Gosal, Cynthia Liu, Zhiming Chen, Osama Mohammed Afzal, Samta Kamboj, Onkar Pandit, Rahul Pal, Lalit Pradhan, Zain Muhammad Mujahid, Massa Baali, Xudong Han, Sondos Mahmoud Bsharat, and 13 others. 2023. [Jais and jais-chat: Arabic-centric foundation and instruction-tuned open generative large language models](https://md.research.elamur.ai/free/en/papers/sengupta-2023-jais-jais-chat-arabic-centric/). Preprint, arXiv:2308.16149. [↗](https://arxiv.org/abs/2308.16149)
65. Siqi Shen, Lajanugen Logeswaran, Moontae Lee, Honglak Lee, Soujanya Poria, and Rada Mihalcea. 2024. Understanding the capabilities and limitations of large language models for cultural commonsense. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), pages 5668–5680. [find](https://scholar.google.com/scholar?q=Siqi+Shen%2C+Lajanugen+Logeswaran%2C+Moontae+Lee%2C+Honglak+Lee%2C+Soujanya+Poria%2C+and+Rada+Mihalcea%2E+2024%2E+Understanding+the+capabilities+and+limitations+of+large+language+models+for+cultural+commonsense%2E+In)
66. Wenxuan Wang, Zhaopeng Tu, Chang Chen, Youliang Yuan, Jen-tse Huang, Wenxiang Jiao, and Michael Lyu. 2024. All languages matter: On the multilingual safety of LLMs. In Findings of the Association for Computational Linguistics: ACL 2024, pages 5865– 5877, Bangkok, Thailand. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Wenxuan+Wang%2C+Zhaopeng+Tu%2C+Chang+Chen%2C+Youliang+Yuan%2C+Jen%2Dtse+Huang%2C+Wenxiang+Jiao%2C+and+Michael+Lyu%2E+2024%2E+All+languages+matter%3A+On+the+multilingual+safety+of+LLMs%2E+In+Findings+of+the+Association+for+)

*Updated 2 November 2026*