levongevorgian/Low-resource-Armenian-NLP
A research project focused on improving NLP for the low-resource Armenian language by analyzing tokenizer inefficiency, training Armenian-specific tokenizers, grafting new vocabulary into the Qwen2.5 model, and performing LoRA fine-tuning to recover performance.
A project to build a Large Language Model (LLM) specifically for the Armenian language from scratch using PyTorch and...
A capstone project evaluating GPT-3.5-Turbo's performance on Armenian language tasks, including extractive QA, multip...
This repository contains the code and data pipelines for a bachelor thesis project on Armenian NLP, specifically focu...
A research project exploring machine learning approaches for detecting loanwords in Armenian and predicting their lan...
A research project exploring how unsupervised pre-training on untranscribed audio from different language families af...
A curated list of Armenian language datasets, corpora, models, and digital resources for NLP and computational lingui...