NaxCyberLead/NigeriaNLP

The definitive hub for low-resource Natural Language Processing (NLP) research for Nigerian languages (Hausa, Yorùbá, Igbo, and beyond). A community-led workshop dedicated to building resources, benchmarks, and models for Africa's most linguistically diverse nation, following the Masakhane ethos. # NigeriaNLP: The Workshop on Natural Language Processing for Nigerian Languages 🌍 Vision and Mission NigeriaNLP is the definitive academic workshop dedicated to advancing Natural Language Processing (NLP) research and resource development for the languages of Nigeria. Modeled after the continental AfricaNLP workshop and rooted in the principles of the Masakhane grassroots movement ("We build together" ), our mission is to empower Africans to shape and own technological advances toward human dignity and equity.   We prioritize research for Africans, by Africans, focusing on collaboration, community ownership, and ethical stewardship of linguistic data.   🇳🇬 The NaijaNLP Imperative Nigeria is one of the most linguistically diverse nations globally, hosting over 500 distinct languages. Despite the demographic size—with major languages like Hausa, Yorùbá, and Igbo spoken by over 175 million people—these languages remain computationally classified as severely low-resource (LRLs). This disparity results in a lack of digitized resources, fragmented data collections, and insufficient publicly available benchmarks.   NigeriaNLP serves as the central hub to address these critical gaps, focusing on high-impact research areas that directly benefit Nigerian linguistic communities. 🔬 Scientific Scope and Call for Papers (CFP) We solicit high-quality, original research and resource papers across a broad range of low-resource computational linguistics topics, with specific emphasis on Nigerian languages. The suggested tracks include, but are not limited to: Low-Resource Language Modeling and Machine Translation (MT): Advancements in Language Modelling (LM) and Translation Aids for Nigerian LRLs.   Techniques leveraging cross-lingual transfer learning (e.g., using specialized models like AfriBERT) for resource-efficient scaling.   Creating and evaluating domain-specific MT models for specialized applications (e.g., education, health).   Multimodal and Acoustic Processing ( …