SEACrowd/sea_madlad

SEA MADLAD is a subset of MADLAD-400 (Multilingual Audited Dataset: Low-resource And Document-level), which is a document-level multilingual dataset based on Common Crawl. SEA MADLAD only filters the language of the "clean" subset, which covers 36 languages indigenous to SEA from 419 languages in total. As a result, some of SEA lang codes aren't available in this version because those belongs to the languages whose decision was to "remove from its clean version" based on MADLAD auditing process. MADLAD uses all snapshots of CommonCrawl available as of August 1, 2022. The primary advantage of this dataset over similar datasets is that it is more multilingual, it is audited and more highly filtered, and it is document-level. The main disadvantage is also its strength -- being more filtered, it may lack the recall needed for some applications.

Languages

ace, akb, ban, bbc, bew, btx, ceb, fil, gor, hil, iba, ilo, ind, jav, kac, khm, kxd, lao, mad, mak, meo, min, mkn, msa, msi, mya, nij, nut, pag, shn, sun, tet, tha, vie, war

Supported Tasks

Self Supervised Pretraining

Dataset Usage

Using datasets library

from datasets import load_dataset
dset = datasets.load_dataset("SEACrowd/sea_madlad", trust_remote_code=True)

Using seacrowd library

# Load the dataset using the default config
dset = sc.load_dataset("sea_madlad", schema="seacrowd")
# Check all available subsets (config names) of the dataset
print(sc.available_config_names("sea_madlad"))
# Load the dataset using a specific config
dset = sc.load_dataset_by_config_name(config_name="<config_name>")

More details on how to load the seacrowd library can be found here.

Dataset Homepage

https://huggingface.co/datasets/allenai/MADLAD-400

Dataset Version

Source: 1.0.0. SEACrowd: 2024.06.20.

Dataset License

Creative Commons Attribution 4.0 (cc-by-4.0)

Citation

If you are using the Sea Madlad dataloader in your work, please cite the following:

@misc{kudugunta2023madlad400,
      title={MADLAD-400: A Multilingual And Document-Level Large Audited Dataset},
      author={Sneha Kudugunta and Isaac Caswell and Biao Zhang and Xavier Garcia and Christopher A. Choquette-Choo and Katherine Lee and Derrick Xin and Aditya Kusupati and Romi Stella and Ankur Bapna and Orhan Firat},
      year={2023},
      eprint={2309.04662},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}


@article{lovenia2024seacrowd,
    title={SEACrowd: A Multilingual Multimodal Data Hub and Benchmark Suite for Southeast Asian Languages}, 
    author={Holy Lovenia and Rahmad Mahendra and Salsabil Maulana Akbar and Lester James V. Miranda and Jennifer Santoso and Elyanah Aco and Akhdan Fadhilah and Jonibek Mansurov and Joseph Marvin Imperial and Onno P. Kampman and Joel Ruben Antony Moniz and Muhammad Ravi Shulthan Habibi and Frederikus Hudi and Railey Montalan and Ryan Ignatius and Joanito Agili Lopo and William Nixon and Börje F. Karlsson and James Jaya and Ryandito Diandaru and Yuze Gao and Patrick Amadeus and Bin Wang and Jan Christian Blaise Cruz and Chenxi Whitehouse and Ivan Halim Parmonangan and Maria Khelli and Wenyu Zhang and Lucky Susanto and Reynard Adha Ryanda and Sonny Lazuardi Hermawan and Dan John Velasco and Muhammad Dehan Al Kautsar and Willy Fitra Hendria and Yasmin Moslem and Noah Flynn and Muhammad Farid Adilazuarda and Haochen Li and Johanes Lee and R. Damanhuri and Shuo Sun and Muhammad Reza Qorib and Amirbek Djanibekov and Wei Qi Leong and Quyet V. Do and Niklas Muennighoff and Tanrada Pansuwan and Ilham Firdausi Putra and Yan Xu and Ngee Chia Tai and Ayu Purwarianti and Sebastian Ruder and William Tjhi and Peerat Limkonchotiwat and Alham Fikri Aji and Sedrick Keh and Genta Indra Winata and Ruochen Zhang and Fajri Koto and Zheng-Xin Yong and Samuel Cahyawijaya},
    year={2024},
    eprint={2406.10118},
    journal={arXiv preprint arXiv: 2406.10118}
}