ArXiv TLDR

Mullosharaf K. Arabov

3 papers · Latest:

Natural Language Processing

Adapting Large Language Models to a Low-Resource Agglutinative Language: A Comparative Study of LoRA and QLoRA for Bashkir

This paper compares LoRA and QLoRA for adapting LLMs to Bashkir, finding QLoRA on 7B models offers a strong quality-cost trade-off.

2605.04948
Natural Language Processing

Natural Language Processing: A Comprehensive Practical Guide from Tokenisation to RLHF

This paper is a practical guide covering the modern NLP pipeline from tokenization to RLHF, emphasizing reproducible research and open-weight models.

2605.03799
Natural Language Processing

Benchmarking Parameter-Efficient Fine-Tuning of Large Language Models for Low-Resource Tajik Text Generation with the Tajik Web Corpus

This paper benchmarks PEFT strategies for LLMs on Tajik, creating the largest Tajik corpus and finding Mistral 7B with QLoRA performs best.

2605.03742

📬 Weekly AI Paper Digest

Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.