June 11, 2024, 10:50 a.m. | Maximilian Schreiner

THE DECODER the-decoder.com


Researchers have introduced a collection of three datasets based on African languages for evaluating large language models. The benchmark set, dubbed IrokoBench, aims to fill a gap in AI research.


The article IrokoBench uncovers a 45% performance gap between English and African languages in LLMs appeared first on THE DECODER.

ai research article artificial intelligence benchmark collection datasets english gap generative-ai language language models languages large language large language models llms performance research researchers set

More from the-decoder.com / THE DECODER

Senior Data Engineer

@ Displate | Warsaw

Sr. Specialist, Research Automation Systems Integrator (Hybrid)

@ MSD | USA - Pennsylvania - West Point

Lead Developer-Process Automation -Python Developer

@ Diageo | Bengaluru Karle Town SEZ

RPA Engineer- Power Automate Desktop, UI Path

@ Baker Hughes | IN-KA-BANGALORE-NEON BUILDING WEST TOWER

Research Fellow (Computer Science (and Engineering)/Electronic Engineering/Applied Mathematics/Perception Sciences)

@ Nanyang Technological University | NTU Main Campus, Singapore

Analista de Ciências de dados II

@ Ingram Micro | BR Link - São Paulo