Reuben fernandes's picture

In a Training Loop 🔄

Reuben fernandes PRO

Reubencf

hugging-science

·

http://reuben-fernandes.xyz/

AI & ML interests

LLM

Recent Activity

reacted to cesear64's post with 🧠 1 day ago

Just published: how we built production Sango (Central African Republic) translation without fine-tuning, parallel corpus, or training compute. The method — vocabulary-augmented prompting with a 581-entry native-speaker-verified lexicon — generalizes to any of the ~2,000 African languages at the same data-poverty level. Recipe, dataset, and code template all included. 📄 Blog: https://huggingface.co/blog/MEYNG/sangoai 📦 Dataset: https://huggingface.co/datasets/MEYNG/sango-vocabulary Would especially value feedback from anyone working on other low-resource African languages — Ewondo, Lingala, Wolof next on our roadmap.

reacted to cesear64's post with 👀 1 day ago

Just published: how we built production Sango (Central African Republic) translation without fine-tuning, parallel corpus, or training compute. The method — vocabulary-augmented prompting with a 581-entry native-speaker-verified lexicon — generalizes to any of the ~2,000 African languages at the same data-poverty level. Recipe, dataset, and code template all included. 📄 Blog: https://huggingface.co/blog/MEYNG/sangoai 📦 Dataset: https://huggingface.co/datasets/MEYNG/sango-vocabulary Would especially value feedback from anyone working on other low-resource African languages — Ewondo, Lingala, Wolof next on our roadmap.

reacted to cesear64's post with 🔥 1 day ago

Just published: how we built production Sango (Central African Republic) translation without fine-tuning, parallel corpus, or training compute. The method — vocabulary-augmented prompting with a 581-entry native-speaker-verified lexicon — generalizes to any of the ~2,000 African languages at the same data-poverty level. Recipe, dataset, and code template all included. 📄 Blog: https://huggingface.co/blog/MEYNG/sangoai 📦 Dataset: https://huggingface.co/datasets/MEYNG/sango-vocabulary Would especially value feedback from anyone working on other low-resource African languages — Ewondo, Lingala, Wolof next on our roadmap.

View all activity

Organizations

authored a paper about 2 months ago

Konkani LLM: Multi-Script Instruction Tuning and Evaluation for a Low-Resource Indian Language

Paper • 2603.23529 • Published Mar 7 • 1