Exploring-the-Linguistic-Efficiency-of-Large-Language-Models-in-Armenian-Discourse
Anahit-N/Exploring-the-Linguistic-Efficiency-of-Large-Language-Models-in-Armenian-Discourse
Summary
A capstone project evaluating GPT-3.5-Turbo's performance on Armenian language tasks, including extractive QA, multiple-choice, knowledge access, and math reasoning. The repository contains four Jupyter notebooks with experiments, an API key file, and a data folder for generated results.