---
title: "Benchmark Data Contamination in — Vilar 2026"
description: "Benchmark data contamination (BDC) happens when test items end up in an LLM's training data."
authors: ["Iriedson Souto Maior de Moraes Vilar", "David Candeia Maia", "João Brunet", "Fabio Morais", "Leandro Balby Marinho"]
year: 2026
venue: "Proceedings of the Fifteenth Language Resources and Evaluation Conference"
source: "https://aclanthology.org/2026.lrec-1.374/"
doi: "10.63317/39wbjvajnh7t"
licence: "CC BY-NC 4.0 (ELRA, LREC proceedings)"
licence_url: "https://creativecommons.org/licenses/by-nc/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/vilar-2026-benchmark-data-contamination-underrepresented-languages/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/vilar-2026-benchmark-data-contamination-underrepresented-languages/)

[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/4/) / Benchmark Data Contamination in Underrepresented Languages: A Comprehensive Analysis Using Brazilian Data

# Benchmark Data Contamination in Underrepresented Languages: A Comprehensive Analysis Using Brazilian Data

**Authors:** Iriedson Souto Maior de Moraes Vilar, David Candeia Maia, João Brunet, Fabio Morais, Leandro Balby Marinho · **Year:** 2026 · **Venue:** Proceedings of the Fifteenth Language Resources and Evaluation Conference · **Licence:** [CC BY-NC 4.0 (ELRA, LREC proceedings)](https://creativecommons.org/licenses/by-nc/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.lrec-1.374/) · [PDF](https://aclanthology.org/2026.lrec-1.374.pdf) · [DOI 10.63317/39wbjvajnh7t](https://doi.org/10.63317/39wbjvajnh7t) · [OpenAlex](https://openalex.org/W7160217549)

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

## Summary
Benchmark data contamination (BDC) happens when test items end up in an LLM's training data. It has been studied mostly for English and Chinese. Low-resource settings have received little attention, even though public exam materials there often feed both training and evaluation sets. The authors present what they describe as the first systematic analysis of BDC for Brazilian Portuguese.

They evaluated 15 specialized and multilingual models, including Sabia-3.1, Tucano-2B4 and GPT-4o, on four multiple-choice test sets, namely HealthQA-BR, the OAB Exams, BLUEX and the ENEM Challenge. For the model side, they added the TS-Guessing protocol to HELM. In this protocol, one wrong answer option is hidden and the model is asked to regenerate it. For the data side, they checked three pretraining corpora with 50-character substring matching, using an extended fork of LLMSanitize.

Both perspectives show contamination. Several models reproduce masked options well above a threshold derived from the data. Stronger memorization is associated with higher benchmark scores. Some corpora contain most of the benchmark material. The results suggest that Portuguese leaderboard scores may be inflated, so accuracy should be reported together with contamination checks.

## Key points
- **Scale:** 15 models, 3 corpora and 4 benchmarks, giving 60 model–benchmark pairs. Benchmark splits come from HELM.
- **Detection signal:** Exact Match (EM) is the primary signal, since it correlates strongly with ROUGE-L F1 (ρ = 0.820). The contamination threshold is EM ≥ 0.02, chosen at an inflection point in the EM distribution.
- **Flagged pairs:** 14 model–benchmark pairs exceed the threshold. Multilingual models memorize more. Small Portuguese-specialized models show little behavioral contamination, even though they were exposed to leaked data.
- **Score inflation:** Accuracy and contamination are positively correlated, with ρ EM = 0.52–0.79 across benchmarks and p \< 0.001 in all cases.
- **Corpus leakage:** Benchmark overlap exceeds 90% in some Portuguese pretraining datasets. Corpora listed among the paper's resources include GigaVerbo, Aroeira and mc4-pt. A replication package is provided.

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

Fields: [Linguistics & languages](https://md.research.elamur.ai/free/en/shelves/linguistics/)

Topics: [Evaluation & benchmarks](https://md.research.elamur.ai/free/en/topics/evaluation-benchmarks/)

## Cite

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

```bibtex
@inproceedings{vilar2026benchmark,
  title = {{Benchmark Data Contamination in Underrepresented Languages: A Comprehensive Analysis Using Brazilian Data}},
  author = {Vilar, Iriedson Souto Maior de Moraes and Maia, David Candeia and Brunet, João and Morais, Fabio and Marinho, Leandro Balby},
  booktitle = {Proceedings of the Fifteenth Language Resources and Evaluation Conference},
  year = {2026},
  doi = {10.63317/39wbjvajnh7t},
  url = {https://doi.org/10.63317/39wbjvajnh7t}
}
```

- [← previous BARRAC: Adaptation of an English Aspect-based Sentiment Analysis Approach for Classification Tasks in Arabic Dialects](https://md.research.elamur.ai/free/en/papers/almutairi-2026-barrac-adaptation-english-aspect-based/)
- [next → Benchmarking Arabic Authorship Attribution and Style Transfer with Large Language Models](https://md.research.elamur.ai/free/en/papers/hamed-2026-benchmarking-arabic-authorship-attribution-style/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/4/)

## References

71 references, 9 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Hugo Abonizio, Thales Sales Almeida, Thiago Laitz, Roseval Malaquias Junior, Giovana Kerche Bonás, Rodrigo Nogueira, and Ramon Pires. 2024. Sabiá-3 Technical Report. arXiv preprint arXiv:2410.12049. — [arXiv:2410.12049](https://arxiv.org/abs/2410.12049) [find](https://scholar.google.com/scholar?q=Hugo+Abonizio%2C+Thales+Sales+Almeida%2C+Thiago+Laitz%2C+Roseval+Malaquias+Junior%2C+Giovana+Kerche+Bon%C3%A1s%2C+Rodrigo+Nogueira%2C+and+Ramon+Pires%2E+2024%2E+Sabi%C3%A1%2D3+Technical+Report%2E+arXiv+preprint+arXiv%3A2410%2E12049%2E+%E2%80%94)
2. Thales Sales Almeida, Thiago Laitz, Giovana K. Bonás, and Rodrigo Nogueira. 2023. BLUEX: A benchmark based on Brazilian leading universities entrance eXams. In Intelligent Systems (BRACIS), pages 337–347. Springer Nature Switzerland. [find](https://scholar.google.com/scholar?q=Thales+Sales+Almeida%2C+Thiago+Laitz%2C+Giovana+K%2E+Bon%C3%A1s%2C+and+Rodrigo+Nogueira%2E+2023%2E+BLUEX%3A+A+benchmark+based+on+Brazilian+leading+universities+entrance+eXams%2E+In+Intelligent+Systems+%28BRACIS%29%2C+pages+337%E2%80%93)
3. Thales Sales Almeida, Rodrigo Nogueira, and Helio Pedrini. 2025. Building high-quality datasets for portuguese llms: From common crawl snapshots to industrial-grade corpora. arXiv preprint arXiv:2509.08824. — [arXiv:2509.08824](https://arxiv.org/abs/2509.08824) [find](https://scholar.google.com/scholar?q=Thales+Sales+Almeida%2C+Rodrigo+Nogueira%2C+and+Helio+Pedrini%2E+2025%2E+Building+high%2Dquality+datasets+for+portuguese+llms%3A+From+common+crawl+snapshots+to+industrial%2Dgrade+corpora%2E+arXiv+preprint+arXiv%3A2509%2E)
4. Edward Bayes, Israel Abebe Azime, Jesujoba O. Alabi, Jonas Kgomo, Tyna Eloundou, Elizabeth Proehl, Kai Chen, Imaan Khadir, Naome A. Etori, Shamsuddeen Hassan Muhammad, Choice Mpanza, Igneciah Pocia Thete, Dietrich Klakow, and David Ifeoluwa Adelani. 2024. Uhura: A benchmark for evaluating scientific… [find](https://scholar.google.com/scholar?q=Edward+Bayes%2C+Israel+Abebe+Azime%2C+Jesujoba+O%2E+Alabi%2C+Jonas+Kgomo%2C+Tyna+Eloundou%2C+Elizabeth+Proehl%2C+Kai+Chen%2C+Imaan+Khadir%2C+Naome+A%2E+Etori%2C+Shamsuddeen+Hassan+Muhammad%2C+Choice+Mpanza%2C+Igneciah+Pocia+Th)
5. Mojca Brglez and Špela Vintar. 2025. From polyester girlfriends to blind mice: Creating the first pragmatics understanding benchmarks for slovene. arXiv preprint arXiv:2510.21575. [find](https://scholar.google.com/scholar?q=Mojca+Brglez+and+%C5%A0pela+Vintar%2E+2025%2E+From+polyester+girlfriends+to+blind+mice%3A+Creating+the+first+pragmatics+understanding+benchmarks+for+slovene%2E+arXiv+preprint+arXiv%3A2510%2E21575%2E)
6. Yupeng Chang, Xu Wang, Jindong Wang, Yuan Wu, Linyi Yang, Kaijie Zhu, Hao Chen, Xiaoyuan Yi, Cunxiang Wang, Yidong Wang, Wei Ye, Yue Zhang, Yi Chang, Philip S. Yu, Qiang Yang, and Xing Xie. 2024. A survey on evaluation of large language models. ACM Transactions on Intelligent Systems and Technology,… [find](https://scholar.google.com/scholar?q=Yupeng+Chang%2C+Xu+Wang%2C+Jindong+Wang%2C+Yuan+Wu%2C+Linyi+Yang%2C+Kaijie+Zhu%2C+Hao+Chen%2C+Xiaoyuan+Yi%2C+Cunxiang+Wang%2C+Yidong+Wang%2C+Wei+Ye%2C+Yue+Zhang%2C+Yi+Chang%2C+Philip+S%2E+Yu%2C+Qiang+Yang%2C+and+Xing+Xie%2E+2024%2E+A+su) [our card](https://md.research.elamur.ai/free/en/papers/chang-2023-survey-evaluation-large-language-models/)
7. Hanjie Chen, Zhouxiang Fang, Yash Singla, and Mark Dredze. 2025. Benchmarking large language models on answering and explaining challenging medical questions. In Proceedings of the 2025 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technolo… [find](https://scholar.google.com/scholar?q=Hanjie+Chen%2C+Zhouxiang+Fang%2C+Yash+Singla%2C+and+Mark+Dredze%2E+2025%2E+Benchmarking+large+language+models+on+answering+and+explaining+challenging+medical+questions%2E+In+Proceedings+of+the+2025+Conference+of+)
8. Mark Chen, Jerry Tworek, Heewoo Jun, Qiming Yuan, Henrique Ponde De Oliveira Pinto, Jared Kaplan, Harri Edwards, Yuri Burda, Nicholas Joseph, Greg Brockman, et al. 2021. Evaluating large language models trained on code. arXiv preprint arXiv:2107.03374. [find](https://scholar.google.com/scholar?q=Mark+Chen%2C+Jerry+Tworek%2C+Heewoo+Jun%2C+Qiming+Yuan%2C+Henrique+Ponde+De+Oliveira+Pinto%2C+Jared+Kaplan%2C+Harri+Edwards%2C+Yuri+Burda%2C+Nicholas+Joseph%2C+Greg+Brockman%2C+et+al%2E+2021%2E+Evaluating+large+language+mode) [our card](https://md.research.elamur.ai/free/en/papers/chen-2021-evaluating-large-language-models-trained/)
9. Karl Cobbe, Vineet Kosaraju, Mohammad Bavarian, Mark Chen, Heewoo Jun, Lukasz Kaiser, Matthias Plappert, Jerry Tworek, Jacob Hilton, Reiichiro Nakano, Christopher Hesse, and John Schulman. 2021. Training verifiers to solve math word problems. arXiv preprint arXiv:2110.14168. [find](https://scholar.google.com/scholar?q=Karl+Cobbe%2C+Vineet+Kosaraju%2C+Mohammad+Bavarian%2C+Mark+Chen%2C+Heewoo+Jun%2C+Lukasz+Kaiser%2C+Matthias+Plappert%2C+Jerry+Tworek%2C+Jacob+Hilton%2C+Reiichiro+Nakano%2C+Christopher+Hesse%2C+and+John+Schulman%2E+2021%2E+Train) [our card](https://md.research.elamur.ai/free/en/papers/cobbe-2021-training-verifiers-solve-math-word/)
10. Nicholas Kluge Corrêa, Sophia Falk, Shiza Fatimah, Aniket Sen, and Nythamar De Oliveira. 2024. TeenyTinyLlama: Open-source tiny language models trained in Brazilian Portuguese. Machine Learning with Applications, 16:100558. [find](https://scholar.google.com/scholar?q=Nicholas+Kluge+Corr%C3%AAa%2C+Sophia+Falk%2C+Shiza+Fatimah%2C+Aniket+Sen%2C+and+Nythamar+De+Oliveira%2E+2024%2E+TeenyTinyLlama%3A+Open%2Dsource+tiny+language+models+trained+in+Brazilian+Portuguese%2E+Machine+Learning+with+A)
11. Nicholas Kluge Corrêa, Aniket Sen, Sophia Falk, and Shiza Fatimah. 2025. Tucano: Advancing Neural Text Generation for Portuguese. Patterns. In Press. [find](https://scholar.google.com/scholar?q=Nicholas+Kluge+Corr%C3%AAa%2C+Aniket+Sen%2C+Sophia+Falk%2C+and+Shiza+Fatimah%2E+2025%2E+Tucano%3A+Advancing+Neural+Text+Generation+for+Portuguese%2E+Patterns%2E+In+Press%2E)
12. Andrew Maranhão Ventura D’addario. 2025. HealthQA-BR: A system-wide benchmark reveals critical knowledge gaps in large language models. arXiv preprint arXiv:2506.21578. — [arXiv:2506.21578](https://arxiv.org/abs/2506.21578) [find](https://scholar.google.com/scholar?q=Andrew+Maranh%C3%A3o+Ventura+D%E2%80%99addario%2E+2025%2E+HealthQA%2DBR%3A+A+system%2Dwide+benchmark+reveals+critical+knowledge+gaps+in+large+language+models%2E+arXiv+preprint+arXiv%3A2506%2E21578%2E+%E2%80%94+%5BarXiv%3A2506%2E21578%5D%28https%3A%2F%2Far)
13. Jasper Dekoninck, Mark Niklas Müller, and Martin Vechev. 2024. Constat: Performance-based contamination detection in large language models. In Advances in Neural Information Processing Systems, volume 37, pages 92420–92464. Curran Associates, Inc. [find](https://scholar.google.com/scholar?q=Jasper+Dekoninck%2C+Mark+Niklas+M%C3%BCller%2C+and+Martin+Vechev%2E+2024%2E+Constat%3A+Performance%2Dbased+contamination+detection+in+large+language+models%2E+In+Advances+in+Neural+Information+Processing+Systems%2C+volume)
14. Pedro Delfino, Bruno Cuconato, Edward Hermann Haeusler, and Alexandre Rademaker. 2017. Passing the Brazilian OAB Exam: Data Preparation and Some Experiments. In Legal Knowledge and Information Systems - JURIX 2017: The Thirtieth Annual Conference, volume 302 of Frontiers in Artificial Intelligence a… [find](https://scholar.google.com/scholar?q=Pedro+Delfino%2C+Bruno+Cuconato%2C+Edward+Hermann+Haeusler%2C+and+Alexandre+Rademaker%2E+2017%2E+Passing+the+Brazilian+OAB+Exam%3A+Data+Preparation+and+Some+Experiments%2E+In+Legal+Knowledge+and+Information+Systems)
15. Chunyuan Deng, Yilun Zhao, Yuzhao Heng, Yitong Li, Jiannan Cao, Xiangru Tang, and Arman Cohan. 2024a. Unveiling the spectrum of data contamination in language models: A survey from detection to remediation. In Findings of the Association for Computational Linguistics: ACL 2024, pages 16078–16092, Ba… [find](https://scholar.google.com/scholar?q=Chunyuan+Deng%2C+Yilun+Zhao%2C+Yuzhao+Heng%2C+Yitong+Li%2C+Jiannan+Cao%2C+Xiangru+Tang%2C+and+Arman+Cohan%2E+2024a%2E+Unveiling+the+spectrum+of+data+contamination+in+language+models%3A+A+survey+from+detection+to+remedi)
16. Chunyuan Deng, Yilun Zhao, Xiangru Tang, Mark Gerstein, and Arman Cohan. 2024b. Investigating data contamination in modern benchmarks for large language models. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Techno… [find](https://scholar.google.com/scholar?q=Chunyuan+Deng%2C+Yilun+Zhao%2C+Xiangru+Tang%2C+Mark+Gerstein%2C+and+Arman+Cohan%2E+2024b%2E+Investigating+data+contamination+in+modern+benchmarks+for+large+language+models%2E+In+Proceedings+of+the+2024+Conference+o) [our card](https://md.research.elamur.ai/free/en/papers/deng-2024-investigating-data-contamination-modern-benchmarks/)
17. Yihong Dong, Xue Jiang, Huanyu Liu, Zhi Jin, Bin Gu, Mengfei Yang, and Ge Li. 2024. Generalization or memorization: Data contamination and trustworthy evaluation for large language models. In Findings of the Association for Computational Linguistics: ACL 2024, pages 12039– 12050, Bangkok, Thailand.… [find](https://scholar.google.com/scholar?q=Yihong+Dong%2C+Xue+Jiang%2C+Huanyu+Liu%2C+Zhi+Jin%2C+Bin+Gu%2C+Mengfei+Yang%2C+and+Ge+Li%2E+2024%2E+Generalization+or+memorization%3A+Data+contamination+and+trustworthy+evaluation+for+large+language+models%2E+In+Findings)
18. Bonaventure F. P. Dossou, Atnafu Lambebo Tonja, Oreen Yousuf, Salomey Osei, Abigail Oppong, Iyanuoluwa Shode, Oluwabusayo Olufunke Awoyomi, and Chris Emezue. 2022. Afrolm: A self-active learning-based multilingual pretrained language model for 23 african languages. In Proceedings of the Third Worksh… [find](https://scholar.google.com/scholar?q=Bonaventure+F%2E+P%2E+Dossou%2C+Atnafu+Lambebo+Tonja%2C+Oreen+Yousuf%2C+Salomey+Osei%2C+Abigail+Oppong%2C+Iyanuoluwa+Shode%2C+Oluwabusayo+Olufunke+Awoyomi%2C+and+Chris+Emezue%2E+2022%2E+Afrolm%3A+A+self%2Dactive+learning%2Dbased)
19. Gabriel Lino Garcia, Pedro Henrique Paiola, Luis Henrique Morelli, Giovani Candido, Arnaldo Cândido Júnior, Danilo Samuel Jodas, Luis Afonso, Ivan Rizzo Guilherme, Bruno Elias Penteado, and João Paulo Papa. 2024. Introducing bode: A fine-tuned large language model for Portuguese prompt-based task. a… — [arXiv:2401.02909](https://arxiv.org/abs/2401.02909) [find](https://scholar.google.com/scholar?q=Gabriel+Lino+Garcia%2C+Pedro+Henrique+Paiola%2C+Luis+Henrique+Morelli%2C+Giovani+Candido%2C+Arnaldo+C%C3%A2ndido+J%C3%BAnior%2C+Danilo+Samuel+Jodas%2C+Luis+Afonso%2C+Ivan+Rizzo+Guilherme%2C+Bruno+Elias+Penteado%2C+and+Jo%C3%A3o+Paulo)
20. Shahriar Golchin and Mihai Surdeanu. 2025. Data Contamination Quiz: A Tool to Detect and Estimate Contamination in Large Language Models. Transactions of the Association for Computational Linguistics, 13:809–830. [find](https://scholar.google.com/scholar?q=Shahriar+Golchin+and+Mihai+Surdeanu%2E+2025%2E+Data+Contamination+Quiz%3A+A+Tool+to+Detect+and+Estimate+Contamination+in+Large+Language+Models%2E+Transactions+of+the+Association+for+Computational+Linguistics%2C)
21. Dan Hendrycks, Collin Burns, Steven Basart, Andy Zou, Mantas Mazeika, Dawn Song, and Jacob Steinhardt. 2021. Measuring massive multitask language understanding. In International Conference on Learning Representations (ICLR). [find](https://scholar.google.com/scholar?q=Dan+Hendrycks%2C+Collin+Burns%2C+Steven+Basart%2C+Andy+Zou%2C+Mantas+Mazeika%2C+Dawn+Song%2C+and+Jacob+Steinhardt%2E+2021%2E+Measuring+massive+multitask+language+understanding%2E+In+International+Conference+on+Learning) [our card](https://md.research.elamur.ai/free/en/papers/hendrycks-2020-measuring-massive-multitask-language-understanding/)
22. Roseval Malaquias Junior, Ramon Pires, Roseli Romero, and Rodrigo Nogueira. 2024. Juru: Legal Brazilian large language model from reputable sources. arXiv preprint arXiv:2403.18140. [find](https://scholar.google.com/scholar?q=Roseval+Malaquias+Junior%2C+Ramon+Pires%2C+Roseli+Romero%2C+and+Rodrigo+Nogueira%2E+2024%2E+Juru%3A+Legal+Brazilian+large+language+model+from+reputable+sources%2E+arXiv+preprint+arXiv%3A2403%2E18140%2E)
23. Mohammad Aflah Khan, Neemesh Yadav, Sarah Masud, and Md. Shad Akhtar. 2025. Quench: Measuring the gap between indic and non-indic contextual general reasoning in llms. In Proceedings of the 31st International Conference on Computational Linguistics, pages 4493–4509, Abu Dhabi, UAE. Association for C… [find](https://scholar.google.com/scholar?q=Mohammad+Aflah+Khan%2C+Neemesh+Yadav%2C+Sarah+Masud%2C+and+Md%2E+Shad+Akhtar%2E+2025%2E+Quench%3A+Measuring+the+gap+between+indic+and+non%2Dindic+contextual+general+reasoning+in+llms%2E+In+Proceedings+of+the+31st+Inter)
24. Mohammed Safi Ur Rahman Khan, Priyam Mehta, Ananth Sankar, Umashankar Kumaravelan, Sumanth Doddapaneni, Suriyaprasaad G, Varun Balan G, Sparsh Jain, Anoop Kunchukuttan, Pratyush Kumar, Raj Dabre, and Mitesh M. Khapra. 2024. Indicllmsuite: A blueprint for creating pre-training and fine-tuning dataset… [find](https://scholar.google.com/scholar?q=Mohammed+Safi+Ur+Rahman+Khan%2C+Priyam+Mehta%2C+Ananth+Sankar%2C+Umashankar+Kumaravelan%2C+Sumanth+Doddapaneni%2C+Suriyaprasaad+G%2C+Varun+Balan+G%2C+Sparsh+Jain%2C+Anoop+Kunchukuttan%2C+Pratyush+Kumar%2C+Raj+Dabre%2C+and+)
25. Julia Kreutzer, Isaac Caswell, Lisa Wang, Ahsan Wahab, Daan Van Esch, Nasanbayar Ulzii-Orshikh, Allahsera Tapo, Nishant Subramani, Artem Sokolov, Claytone Sikasote, et al. 2022. Quality at a glance: An audit of web-crawled multilingual datasets. Transactions of the Association for Computational Ling… [find](https://scholar.google.com/scholar?q=Julia+Kreutzer%2C+Isaac+Caswell%2C+Lisa+Wang%2C+Ahsan+Wahab%2C+Daan+Van+Esch%2C+Nasanbayar+Ulzii%2DOrshikh%2C+Allahsera+Tapo%2C+Nishant+Subramani%2C+Artem+Sokolov%2C+Claytone+Sikasote%2C+et+al%2E+2022%2E+Quality+at+a+glance%3A+A) [our card](https://md.research.elamur.ai/free/en/papers/kreutzer-2021-quality-glance-audit-web-crawled/)
26. Celio Larcher, Marcos Piau, Paulo Finardi, Pedro Gengo, Piero Esposito, and Vinicius Caridá. 2023. Cabrita: closing the gap for foreign languages. arXiv preprint arXiv:2308.11878. — [arXiv:2308.11878](https://arxiv.org/abs/2308.11878) [find](https://scholar.google.com/scholar?q=Celio+Larcher%2C+Marcos+Piau%2C+Paulo+Finardi%2C+Pedro+Gengo%2C+Piero+Esposito%2C+and+Vinicius+Carid%C3%A1%2E+2023%2E+Cabrita%3A+closing+the+gap+for+foreign+languages%2E+arXiv+preprint+arXiv%3A2308%2E11878%2E+%E2%80%94+%5BarXiv%3A2308%2E11878%5D)
27. Md Tahmid Rahman Laskar, Sawsan Alqahtani, M Saiful Bari, Mizanur Rahman, Mohammad Abdullah Matin Khan, Haidar Khan, Israt Jahan, Amran Bhuiyan, Chee Wei Tan, Md Rizwan Parvez, Enamul Hoque, Shafiq Joty, and Jimmy Huang. 2024. A systematic survey and critical review on evaluating large language mode… [find](https://scholar.google.com/scholar?q=Md+Tahmid+Rahman+Laskar%2C+Sawsan+Alqahtani%2C+M+Saiful+Bari%2C+Mizanur+Rahman%2C+Mohammad+Abdullah+Matin+Khan%2C+Haidar+Khan%2C+Israt+Jahan%2C+Amran+Bhuiyan%2C+Chee+Wei+Tan%2C+Md+Rizwan+Parvez%2C+Enamul+Hoque%2C+Shafiq+Jo)
28. Zihao Li, Yucheng Shi, Zirui Liu, Fan Yang, Ali Payani, Ninghao Liu, and Mengnan Du. 2025. Language Ranker: A metric for quantifying LLM performance across high and low-resource languages. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 39, pages 28186–28194. AAAI Press. [find](https://scholar.google.com/scholar?q=Zihao+Li%2C+Yucheng+Shi%2C+Zirui+Liu%2C+Fan+Yang%2C+Ali+Payani%2C+Ninghao+Liu%2C+and+Mengnan+Du%2E+2025%2E+Language+Ranker%3A+A+metric+for+quantifying+LLM+performance+across+high+and+low%2Dresource+languages%2E+In+Proceedi)
29. Percy Liang, Rishi Bommasani, Tony Lee, Dimitris Tsipras, Dilara Soylu, Michihiro Yasunaga, Yian Zhang, Deepak Narayanan, Yuhuai Wu, Ananya Kumar, et al. 2023. Holistic evaluation of language models. Transactions on Machine Learning Research. [find](https://scholar.google.com/scholar?q=Percy+Liang%2C+Rishi+Bommasani%2C+Tony+Lee%2C+Dimitris+Tsipras%2C+Dilara+Soylu%2C+Michihiro+Yasunaga%2C+Yian+Zhang%2C+Deepak+Narayanan%2C+Yuhuai+Wu%2C+Ananya+Kumar%2C+et+al%2E+2023%2E+Holistic+evaluation+of+language+models%2E+)
30. Thiago Lira, Flávio Cação, Cinthia Souza, João Valentini, Edson Bollis, Otavio Oliveira, Renato Almeida, Marcio Magalhães, Katia Poloni, Andre Oliveira, et al. 2025. Aroeira: A curated corpus for the portuguese language with a large number of tokens. In Intelligent Systems, volume 15412 of Lecture N… [find](https://scholar.google.com/scholar?q=Thiago+Lira%2C+Fl%C3%A1vio+Ca%C3%A7%C3%A3o%2C+Cinthia+Souza%2C+Jo%C3%A3o+Valentini%2C+Edson+Bollis%2C+Otavio+Oliveira%2C+Renato+Almeida%2C+Marcio+Magalh%C3%A3es%2C+Katia+Poloni%2C+Andre+Oliveira%2C+et+al%2E+2025%2E+Aroeira%3A+A+curated+corpus+for+the+)
31. Shamsuddeen Hassan Muhammad et al. 2023. Afrisenti: A twitter sentiment analysis benchmark for african languages. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 13968–13981, Singapore. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Shamsuddeen+Hassan+Muhammad+et+al%2E+2023%2E+Afrisenti%3A+A+twitter+sentiment+analysis+benchmark+for+african+languages%2E+In+Proceedings+of+the+2023+Conference+on+Empirical+Methods+in+Natural+Language+Process) [our card](https://md.research.elamur.ai/free/en/papers/muhammad-2023-afrisenti-twitter-sentiment-analysis-benchmark/)
32. Ng, others, and Leslie Teo. 2025. Sea-lion: Southeast asian languages in one network. In Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics, pages 512–526, Mumbai, I… [find](https://scholar.google.com/scholar?q=Ng%2C+others%2C+and+Leslie+Teo%2E+2025%2E+Sea%2Dlion%3A+Southeast+asian+languages+in+one+network%2E+In+Proceedings+of+the+14th+International+Joint+Conference+on+Natural+Language+Processing+and+the+4th+Conference+of)
33. OpenAI. 2023. GPT-4 technical report. preprint arXiv:2303.08774. arXiv Michał Perełkiewicz and Rafał Poświata. 2024. A review of the challenges with massive web-mined corpora used in large language models pre-training. In Artificial Intelligence and Soft Computing, volume 14678 of Lecture Notes in C… [find](https://scholar.google.com/scholar?q=OpenAI%2E+2023%2E+GPT%2D4+technical+report%2E+preprint+arXiv%3A2303%2E08774%2E+arXiv+Micha%C5%82+Pere%C5%82kiewicz+and+Rafa%C5%82+Po%C5%9Bwiata%2E+2024%2E+A+review+of+the+challenges+with+massive+web%2Dmined+corpora+used+in+large+language+mo) [our card](https://md.research.elamur.ai/free/en/papers/openai-2023-gpt-4-technical-report/)
34. Ramon Pires, Hugo Abonizio, Thales Sales Almeida, and Rodrigo Nogueira. 2023. Sabiá: Portuguese large language models. In Intelligent Systems, pages 226–240, Cham. Springer Nature Switzerland. [find](https://scholar.google.com/scholar?q=Ramon+Pires%2C+Hugo+Abonizio%2C+Thales+Sales+Almeida%2C+and+Rodrigo+Nogueira%2E+2023%2E+Sabi%C3%A1%3A+Portuguese+large+language+models%2E+In+Intelligent+Systems%2C+pages+226%E2%80%93240%2C+Cham%2E+Springer+Nature+Switzerland%2E)
35. Libo Qin, Qiguang Chen, Yuhang Zhou, Zhi Chen, Yinghui Li, Lizi Liao, Min Li, Wanxiang Che, and Philip S Yu. 2025. A survey of multilingual large language models. Patterns, 6(1):101118. [find](https://scholar.google.com/scholar?q=Libo+Qin%2C+Qiguang+Chen%2C+Yuhang+Zhou%2C+Zhi+Chen%2C+Yinghui+Li%2C+Lizi+Liao%2C+Min+Li%2C+Wanxiang+Che%2C+and+Philip+S+Yu%2E+2025%2E+A+survey+of+multilingual+large+language+models%2E+Patterns%2C+6%281%29%3A101118%2E)
36. Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. 2019. Language models are unsupervised multitask learners. OpenAI Blog. Accessed on: 2024-10-25. [find](https://scholar.google.com/scholar?q=Alec+Radford%2C+Jeffrey+Wu%2C+Rewon+Child%2C+David+Luan%2C+Dario+Amodei%2C+and+Ilya+Sutskever%2E+2019%2E+Language+models+are+unsupervised+multitask+learners%2E+OpenAI+Blog%2E+Accessed+on%3A+2024%2D10%2D25%2E)
37. Mathieu Ravaut, Bosheng Ding, Fangkai Jiao, Hailin Chen, Xingxuan Li, Ruochen Zhao, Chengwei Qin, Caiming Xiong, and Shafiq Joty. 2024. How much are llms contaminated? a comprehensive survey and the llmsanitize library. arXiv preprint arXiv:2404.00699. [find](https://scholar.google.com/scholar?q=Mathieu+Ravaut%2C+Bosheng+Ding%2C+Fangkai+Jiao%2C+Hailin+Chen%2C+Xingxuan+Li%2C+Ruochen+Zhao%2C+Chengwei+Qin%2C+Caiming+Xiong%2C+and+Shafiq+Joty%2E+2024%2E+How+much+are+llms+contaminated%3F+a+comprehensive+survey+and+the+l)
38. Martin Riddell, Ansong Ni, and Arman Cohan. 2024. Quantifying contamination in evaluating code generation capabilities of language models. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 14116–14137, Bangkok, Thailand. Associa… [find](https://scholar.google.com/scholar?q=Martin+Riddell%2C+Ansong+Ni%2C+and+Arman+Cohan%2E+2024%2E+Quantifying+contamination+in+evaluating+code+generation+capabilities+of+language+models%2E+In+Proceedings+of+the+62nd+Annual+Meeting+of+the+Association+)
39. Oscar Sainz, Jon Ander Campos, Iker García-Ferrero, Julen Etxaniz, Oier Lopez de Lacalle, and Eneko Agirre. 2023. NLP evaluation in trouble: On the need to measure LLM data contamination for each benchmark. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 10776–10787,… [find](https://scholar.google.com/scholar?q=Oscar+Sainz%2C+Jon+Ander+Campos%2C+Iker+Garc%C3%ADa%2DFerrero%2C+Julen+Etxaniz%2C+Oier+Lopez+de+Lacalle%2C+and+Eneko+Agirre%2E+2023%2E+NLP+evaluation+in+trouble%3A+On+the+need+to+measure+LLM+data+contamination+for+each+benc) [our card](https://md.research.elamur.ai/free/en/papers/sainz-2023-nlp-evaluation-trouble-need-measure/)
40. Igor Cataneo Silveira and Denis Deratani Mauá. 2018. Advances in automatically solving the enem. In 2018 7th Brazilian Conference on Intelligent Systems (BRACIS), pages 43–48. [find](https://scholar.google.com/scholar?q=Igor+Cataneo+Silveira+and+Denis+Deratani+Mau%C3%A1%2E+2018%2E+Advances+in+automatically+solving+the+enem%2E+In+2018+7th+Brazilian+Conference+on+Intelligent+Systems+%28BRACIS%29%2C+pages+43%E2%80%9348%2E)
41. Dingjie Song, Sicheng Lai, Mingxuan Wang, Shunian Chen, Lichao Sun, and Benyou Wang. 2025. Both text and images leaked! a systematic analysis of data contamination in multimodal LLMs. In ICML 2025 Workshop on Data in Generative Models - The Bad, the Ugly, and the Greats. [find](https://scholar.google.com/scholar?q=Dingjie+Song%2C+Sicheng+Lai%2C+Mingxuan+Wang%2C+Shunian+Chen%2C+Lichao+Sun%2C+and+Benyou+Wang%2E+2025%2E+Both+text+and+images+leaked%21+a+systematic+analysis+of+data+contamination+in+multimodal+LLMs%2E+In+ICML+2025+Wor)
42. Yifan Sun, Han Wang, Dongbai Li, Gang Wang, and Huan Zhang. 2025. The emperor’s new clothes in benchmarking? a rigorous examination of mitigation strategies for LLM benchmark data contamination. In Proceedings of the 42nd International Conference on Machine Learning (ICML), Vancouver, Canada. PMLR.… [find](https://scholar.google.com/scholar?q=Yifan+Sun%2C+Han+Wang%2C+Dongbai+Li%2C+Gang+Wang%2C+and+Huan+Zhang%2E+2025%2E+The+emperor%E2%80%99s+new+clothes+in+benchmarking%3F+a+rigorous+examination+of+mitigation+strategies+for+LLM+benchmark+data+contamination%2E+In+Pr)
43. Brian Thompson, Mehak Dhaliwal, Peter Frisch, Tobias Domhan, and Marcello Federico. 2024. A shocking amount of the web is machine translated: Insights from multi-way parallelism. In Findings of the Association for Computational Linguistics: ACL 2024, pages 1763–1775, Bangkok, Thailand. Association f… [find](https://scholar.google.com/scholar?q=Brian+Thompson%2C+Mehak+Dhaliwal%2C+Peter+Frisch%2C+Tobias+Domhan%2C+and+Marcello+Federico%2E+2024%2E+A+shocking+amount+of+the+web+is+machine+translated%3A+Insights+from+multi%2Dway+parallelism%2E+In+Findings+of+the+As)
44. Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yasmine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava, Shruti Bhosale, et al. 2023. Llama 2: Open Foundation and Fine-Tuned Chat Models. arXiv preprint arXiv:2307.09288. [find](https://scholar.google.com/scholar?q=Hugo+Touvron%2C+Louis+Martin%2C+Kevin+Stone%2C+Peter+Albert%2C+Amjad+Almahairi%2C+Yasmine+Babaei%2C+Nikolay+Bashlykov%2C+Soumya+Batra%2C+Prajjwal+Bhargava%2C+Shruti+Bhosale%2C+et+al%2E+2023%2E+Llama+2%3A+Open+Foundation+and+Fi)
45. Elif Ecem Umutlu, Ayse Aysu Cengiz, Ahmet Kaan Sever, Seyma Erdem, Burak Aytan, Busra Tufan, Abdullah Topraksoy, Esra Darıcı, and Cagri Toraman. 2025. Evaluating the quality of benchmark datasets for low-resource languages: A case study on turkish. In Proceedings of the Fourth Workshop on Generation… [find](https://scholar.google.com/scholar?q=Elif+Ecem+Umutlu%2C+Ayse+Aysu+Cengiz%2C+Ahmet+Kaan+Sever%2C+Seyma+Erdem%2C+Burak+Aytan%2C+Busra+Tufan%2C+Abdullah+Topraksoy%2C+Esra+Dar%C4%B1c%C4%B1%2C+and+Cagri+Toraman%2E+2025%2E+Evaluating+the+quality+of+benchmark+datasets+for+)
46. Cheng Xu, Shuhao Guan, Derek Greene, and M-Tahar Kechadi. 2024a. Benchmark data contamination of large language models: A survey. arXiv preprint arXiv:2406.04244. [find](https://scholar.google.com/scholar?q=Cheng+Xu%2C+Shuhao+Guan%2C+Derek+Greene%2C+and+M%2DTahar+Kechadi%2E+2024a%2E+Benchmark+data+contamination+of+large+language+models%3A+A+survey%2E+arXiv+preprint+arXiv%3A2406%2E04244%2E)
47. Ruijie Xu, Zengzhi Wang, Run-Ze Fan, and Pengfei Liu. 2024b. Benchmarking benchmark leakage in large language models. arXiv preprint arXiv:2404.18824. [find](https://scholar.google.com/scholar?q=Ruijie+Xu%2C+Zengzhi+Wang%2C+Run%2DZe+Fan%2C+and+Pengfei+Liu%2E+2024b%2E+Benchmarking+benchmark+leakage+in+large+language+models%2E+arXiv+preprint+arXiv%3A2404%2E18824%2E)
48. Fei Yuan, Shuai Yuan, Zhiyong Wu, and Lei Li. 2024. How vocabulary sharing facilitates multilingualism in LLaMA? In Findings of the Association for Computational Linguistics: ACL 2024, pages 12111–12130, Bangkok, Thailand. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Fei+Yuan%2C+Shuai+Yuan%2C+Zhiyong+Wu%2C+and+Lei+Li%2E+2024%2E+How+vocabulary+sharing+facilitates+multilingualism+in+LLaMA%3F+In+Findings+of+the+Association+for+Computational+Linguistics%3A+ACL+2024%2C+pages+12111%E2%80%93121)
49. Chen Zhang, Mingxu Tao, Zhiyuan Liao, and Yansong Feng. 2025. Milic-eval: Benchmarking multilingual llms for china’s minority languages. In Findings of the Association for Computational Linguistics: ACL 2025, pages 11086–11102, Vienna, Austria. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Chen+Zhang%2C+Mingxu+Tao%2C+Zhiyuan+Liao%2C+and+Yansong+Feng%2E+2025%2E+Milic%2Deval%3A+Benchmarking+multilingual+llms+for+china%E2%80%99s+minority+languages%2E+In+Findings+of+the+Association+for+Computational+Linguistics%3A+A)
50. Qin Zhu, Qingyuan Cheng, Runyu Peng, Xiaonan Li, Tengxiao Liu, Ru Peng, Xipeng Qiu, and Xuanjing Huang. 2024a. Inference-time decontamination: Reusing leaked benchmarks for large language model evaluation. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 9113–9129, Mia… [find](https://scholar.google.com/scholar?q=Qin+Zhu%2C+Qingyuan+Cheng%2C+Runyu+Peng%2C+Xiaonan+Li%2C+Tengxiao+Liu%2C+Ru+Peng%2C+Xipeng+Qiu%2C+and+Xuanjing+Huang%2E+2024a%2E+Inference%2Dtime+decontamination%3A+Reusing+leaked+benchmarks+for+large+language+model+evalua)
51. Wenhong Zhu, Hongkun Hao, Zhiwei He, Yun-Ze Song, Yumeng Zhang, Hanxu Hu, Yiran Wei, Rui Wang, and Hongyuan Lu. 2024b. CLEAN-EVAL: Clean evaluation on contaminated large language models. In Findings of the Association for Computational Linguistics: NAACL 2024, pages 835–847, Mexico City, Mexico. Ass… [find](https://scholar.google.com/scholar?q=Wenhong+Zhu%2C+Hongkun+Hao%2C+Zhiwei+He%2C+Yun%2DZe+Song%2C+Yumeng+Zhang%2C+Hanxu+Hu%2C+Yiran+Wei%2C+Rui+Wang%2C+and+Hongyuan+Lu%2E+2024b%2E+CLEAN%2DEVAL%3A+Clean+evaluation+on+contaminated+large+language+models%2E+In+Findings+o)
52. C. G. Camilo-Junior, S. S. T. Oliveira, L. A. Pereira, M. Amadeus, R. Scotti, D. Fazzioni, A. M. A. Novais, and S. A. A. Jordão. 2025. GAIA: An Open Language Model for Brazilian Portuguese. [https://huggingface.co/CEIA-UFG/Gemma-3-Gaia-PT-BR-4b-it](https://huggingface.co/CEIA-UFG/Gemma-3-Gaia-PT-BR-4b-it). Accessed at: 2025-10-17. [↗](https://huggingface.co/CEIA-UFG/Gemma-3-Gaia-PT-BR-4b-it)
53. Eduardo Garcia. 2023. mc4-pt. [https://huggingface.co/datasets/eduagarcia/mc4-pt](https://huggingface.co/datasets/eduagarcia/mc4-pt). Accessed at: 2025-10-17. [↗](https://huggingface.co/datasets/eduagarcia/mc4-pt)
54. Eduardo Garcia. 2024a. enem_challenge. [https://huggingface.co/datasets/eduagarcia/enem_challenge](https://huggingface.co/datasets/eduagarcia/enem_challenge). Accessed at: 2025-10-17. [↗](https://huggingface.co/datasets/eduagarcia/enem_challenge)
55. Eduardo Garcia. 2024b. oab_exams. [https://huggingface.co/datasets/Accessed](https://huggingface.co/datasets/Accessed) at: eduagarcia/oab_exams. 2025-10-17. [↗](https://huggingface.co/datasets/Accessed)
56. Google. 2025. Gemma-3-12B-it. [https://huggingface.co/google/gemma-3-12b-it](https://huggingface.co/google/gemma-3-12b-it). Accessed at: 2025-10-17. [↗](https://huggingface.co/google/gemma-3-12b-it)
57. Granite Team, IBM. 2025. Granite-3.3-8B-Instruct. [https://huggingface.co/ibm-granite/granite-3](https://huggingface.co/ibm-granite/granite-3). Accessed at: 2025-10-3-8b-instruct. 17. [↗](https://huggingface.co/ibm-granite/granite-3)
58. Itaú Unibanco. 2024. Aroeira. [https://huggingface.co/datasets/](https://huggingface.co/datasets/) Accessed at: Itau-Unibanco/aroeira. 2025-10-17. [↗](https://huggingface.co/datasets/)
59. Larxel. 2025. healthqa-br. [https://huggingface.co/datasets/Larxel/](https://huggingface.co/datasets/Larxel/) healthqa-br. Accessed at: 2025-10-17. [↗](https://huggingface.co/datasets/Larxel/)
60. Maritaca AI. 2023. Sabia-7B. [https://huggingface.co/maritaca-ai/](https://huggingface.co/maritaca-ai/) sabia-7b. Accessed at: 2025-10-17. [↗](https://huggingface.co/maritaca-ai/)
61. Maritaca AI. 2025. Sabia 3.1. [https://docs.maritaca.ai/api/pt/list-models](https://docs.maritaca.ai/api/pt/list-models). Accessed at: 2025-10-17. [↗](https://docs.maritaca.ai/api/pt/list-models)
62. Meta AI. 2024. Llama 3.1 8B Instruct. [https://huggingface.co/meta-llama/Accessed](https://huggingface.co/meta-llama/Accessed) at: Llama-3.1-8B-Instruct. 2025-10-17. [↗](https://huggingface.co/meta-llama/Accessed)
63. Meta AI. 2025. Llama-4-Maverick-17B-Instruct. [https://huggingface.co/meta-llama/Ac-Llama-4-Maverick-17B-Instruct](https://huggingface.co/meta-llama/Ac-Llama-4-Maverick-17B-Instruct). cessed at: 2025-10-17. Model referred to as Llama-4-Maverick-17B in table 2. [↗](https://huggingface.co/meta-llama/Ac-Llama-4-Maverick-17B-Instruct)
64. Microsoft. 2024. Phi-4. [https://huggingface.co/microsoft/phi-4](https://huggingface.co/microsoft/phi-4). Accessed at: 2025-10-17. Model referred to as Phi-4 (mid) in table 2. [↗](https://huggingface.co/microsoft/phi-4)
65. Mistral AI. 2024. Ministral-8B-Instruct. [https://huggingface.co/mistralai/Ministral-8B-Instruct-2410](https://huggingface.co/mistralai/Ministral-8B-Instruct-2410). Accessed at: 2025-10-17. Closest official model to Ministral-8B-Instr. in table 2. [↗](https://huggingface.co/mistralai/Ministral-8B-Instruct-2410)
66. OpenAI. 2024. GPT-4o. https://platform. Ac-openai.com/docs/models/gpt-4o. cessed at: 2025-10-17. Model referred to as GPT-4o-2024-11-20 in table 2. [find](https://scholar.google.com/scholar?q=OpenAI%2E+2024%2E+GPT%2D4o%2E+https%3A%2F%2Fplatform%2E+Ac%2Dopenai%2Ecom%2Fdocs%2Fmodels%2Fgpt%2D4o%2E+cessed+at%3A+2025%2D10%2D17%2E+Model+referred+to+as+GPT%2D4o%2D2024%2D11%2D20+in+table+2%2E)
67. Portuguese Benchmark Datasets. 2023. BLUEX. [https://huggingface.co/datasets/portuguese-benchmark-datasets/BLUEX](https://huggingface.co/datasets/portuguese-benchmark-datasets/BLUEX). Accessed at: 2025-10-17. [↗](https://huggingface.co/datasets/portuguese-benchmark-datasets/BLUEX)
68. PORTULAN. 2024. Gervásio-7B. [https://huggingface.co/PORTULAN/gervasio-7b-portuguese-ptbr-decoder](https://huggingface.co/PORTULAN/gervasio-7b-portuguese-ptbr-decoder). Accessed at: 2025-10-17. [↗](https://huggingface.co/PORTULAN/gervasio-7b-portuguese-ptbr-decoder)
69. Qwen Team, Alibaba Cloud. 2025. Qwen3-https://huggingface.co/Qwen/8B. Qwen3-8B. Accessed at: 2025-10-17. Model referred to as Qwen3-8B in table 2. recogna-nlp. 2024. Bode-13B-https://Alpaca-PT-BR. huggingface.co/recogna-nlp/ bode-13b-alpaca-pt-br-no-peft. Accessed at: 2025-10-17. [find](https://scholar.google.com/scholar?q=Qwen+Team%2C+Alibaba+Cloud%2E+2025%2E+Qwen3%2Dhttps%3A%2F%2Fhuggingface%2Eco%2FQwen%2F8B%2E+Qwen3%2D8B%2E+Accessed+at%3A+2025%2D10%2D17%2E+Model+referred+to+as+Qwen3%2D8B+in+table+2%2E+recogna%2Dnlp%2E+2024%2E+Bode%2D13B%2Dhttps%3A%2F%2FAlpaca%2DPT%2DBR%2E+hug)
70. TucanoBR. 2024a. GigaVerbo. [https://huggingface.co/datasets/TucanoBR/](https://huggingface.co/datasets/TucanoBR/) GigaVerbo. Accessed at: 2025-10-17. [↗](https://huggingface.co/datasets/TucanoBR/)
71. TucanoBR. 2024b. Tucano-2B4. [https://huggingface.co/TucanoBR/Tucano-2b4](https://huggingface.co/TucanoBR/Tucano-2b4). Accessed at: 2025-10-17. [↗](https://huggingface.co/TucanoBR/Tucano-2b4)