NLP702 Project - Instructing Generative Models for the Task of Arabic Question Answering (Code to be Uploaded)
This course project focuses on evaluating dif- ferent models for question answering in Ara- bic, specifically using AraGPT2 and mGPT as the models for evaluation. The project em- ploys manual evaluation, following a similar approach as the self-instruct paper, to rank the quality of the models’ output. The evaluation aims to assess the performance of AraGPT2 and mGPT in generating accurate and contex- tually relevant answers to questions in Arabic. The results of the evaluation will provide in- sights into the strengths and weaknesses of these models for question answering in Ara- bic, and contribute to the body of knowledge in natural language processing and Arabic lan- guage processing.