I'm an ML Research Engineer at the Vella.ai, I'm interested in multilingual and multimodal systems, deep learning, and probabilistic methods. I completed my BS in Computer Science from Kotebe University in 2023. You can reach out to me at henokb2124 at gmail dotcom.
All Languages Matter: Evaluating LMMs on Culturally Diverse 100 Languages
Ashmal Vayani, Dinura Dissanayake, {...}, Henok Biadglign Ademtew, {...}, Thamar Solorio, Monojit Choudhury, Ivan Laptev, Mubarak Shah, Salman Khan, Fahad Khan
Arxiv preprint
ProverbEval: Exploring LLM Evaluation Challenges for Low-resource Language Understanding
Israel Abebe Azime, Atnafu Lambebo Tonja, Tadesse Destaw Belay, Yonas Chanie, Bontu Fufa Balcha, Negasi Haile Abadi, Henok Biadglign Ademtew, Mulubrhan Abebe Nerea, Debela Desalegn Yadeta, Derartu Dagne Geremew, Philipp Slusallek, Thamar Solorio, Dietrich Klakow
Arxiv preprint
Many-Shot In-Context Learning: A Solution or Just an Enhancement for Low-Resource NLP?
Henok Biadglign Ademtew, Mikiyas Girma Birbo
Black in AI workshop @Neurips 2024
CVQA: Culturally-diverse Multilingual Visual Question Answering Benchmark
David Romero, Chenyang Lyu, {...}, Henok Biadglign Ademtew,{...}, Oana Ignat, Rada Mihalcea, Thamar Solorio, Alham Fikri Aji
Neurips 2024 Datasets and Benchmarks 🎖Oral
AGE: Amharic, Ge’ez and English Parallel Dataset
Henok Biadglign Ademtew, Mikiyas Girma Birbo
Proceedings of the The Seventh Workshop on Technologies for Machine Translation of Low-Resource Languages (LoResMT 2024), ACL 2024
This paper was also presented in AfricaNLP at ICLR 2024
I collaborated in the development of AYA an initiative to accelerate multilingual AI progress. It is part of Cohere for AI, a project to make AI more accessible. I lead the Amharic language team as the ambassador and main contributor. Aya is now an open source SOTA multilingual dataset. It is now vailable on Huggingface🤗 Datasets! Cohere's and C4AI new open source model, ⌘R+, was trained with our work! The model is now available on Huggingface🤗 Models!