SEACrowd is a community dedicated to bridging the gap between multilingual AI and Southeast Asian AI and enhancing the quality of AI research and researchers in the region. This initiative is co-founded by Southeast Asian researchers and practitioners from diverse institutions.

See what indigenous and non-indigenous languages are under our study here.

Our members have been engaged in AI research in Southeast Asia since 2016. Collectively, our research efforts have transformed the landscape of AI in the region, providing other AI practitioners with access to SEA resources, models, and insights, thereby advancing AI development in Southeast Asia.

Explore the complete list of our contributors' SEA-related publications!

• 11/2023 to 06/2024. SEACrowd: A Multilingual Multimodal Data Hub and Benchmark Suite for Southeast Asian Languages. Accepted in EMNLP 2024. [Announcement]

We will launch our new project in early Nov 2024 (11/2024). Stay tuned!

SIGSEA aims to change that by promoting research, collaboration, and sharing updates on Southeast Asian NLP. This way, in the future, we also can hold our own SEA workshops & events in ACL conferences! 💪

We're collecting expressions of interest for membership. As a member, you'll receive regular updates on research, events, and opportunities in the region.

Everyone can join (no need for ACL membership). Sign up today to join SIGSEA via this form! 🫶

On-going since 08/2024.

This program targets early-career AI enthusiasts from underserved Southeast Asian communities, who are looking to gain their first substantial research experience. Many face challenges such as limited access to research tools, mentorship, and AI developments.

Our program addresses these gaps by providing research problems for participants to solve in small teams, guided by experienced mentors. Through hands-on projects and learning key concepts, participants work toward writing a publication for top AI conferences like ACL. The program also emphasizes critical thinking, collaboration, and academic writing to prepare participants for success in AI research.

Southeast Asia is home to more than 1,000 native languages. Nevertheless, Southeast Asian NLP, vision-language, and speech processing is underrepresented in the research community, and one of the reasons is the lack of access to public datasets (Aji et al., 2022). It is essential to greatly increase the accessibility of SEA datasets, promote research in SEA languages and cultures, as well as build more AI models that represent SEA.

We consolidate a comprehensive resource hub, SEACrowd Catalogue that fills the resource gap in nearly 1,000 Southeast Asian (SEA) languages across three modalities. Find their standardized dataloaders in the SEACrowd Data Hub.

Furthermore, we also build the SEACrowd Benchmarks to assess the quality of AI models on 36 indigenous languages across 13 tasks, offering valuable insights into the current AI landscape in SEA. Furthermore, we propose strategies to facilitate greater AI advancements, maximizing potential utility and resource equity for the future of AI in SEA.

🤔 More info on our website.

Everyone can join and contribute to this initiative.

You can submit detailed metadata for existing datasets through this form. You will provide important information such as data license, size, language and dialect, annotation method, and so on. The approved datasheets as well as under review datasheets will show up and indexed in SEACrowd Catalogue (web/csv).

From the approved datasheets from the previous task, you can help us build HuggingFace’s dataset dataloader to ensure that all datasets in SEACrowd are standardized in terms of formatting. You can take a look at the dataloader guide and examples in SEACrowd Data Hub. We will also ping the taken dataloader issues after 2 weeks of inactivity in case there's any trouble.

Definitely. Please feel free to ask in #discussion-forum so we can help you. 😉

Join our discussions to keep in touch and be the first to receive the updates!

• #general
• Mailing list
• Email: [email protected]
• Repositories: GitHub, HuggingFace

If you are using any resources from SEACrowd, including datasheets, dataloaders, code, etc., please cite the following publication:

@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}
}