Logo Lanfrica

Prompt engineering on large language models (LLMs) in low-resourced language setting

Domain:

natural language processing

Record type:

datasetpaper
Creator:
Pit
Host:avatar

The attached dataset has all the information in regard to Large Language Models (LLMs) being used in low-resource language settings. There are figures depicting the prompt structure used to improve translation of LLMs translation of low-resourced languages, more specifically, low-resourced, morphologically rich African languages. Furthermore, there are figures that depict the results of various experiments conducted in efforts to improve LLMs' translation of low-resourced languages, and there are clean evaluation datasets that can be used by other researchers to evaluate their LLMs in low-resourced language settings.