Enregistré dans:
Détails bibliographiques
Auteurs principaux: Fasha, Mohammed, Hammo, Bassam, Sowan, Bilal, Barham, Husam, Nsour, Esam
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:https://arxiv.org/abs/2601.17364
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866911396280664064
author Fasha, Mohammed
Hammo, Bassam
Sowan, Bilal
Barham, Husam
Nsour, Esam
author_facet Fasha, Mohammed
Hammo, Bassam
Sowan, Bilal
Barham, Husam
Nsour, Esam
contents This study uses Jordanian law as a case study to explore the fine-tuning of the Llama-3.1 large language model for Arabic question-answering. Two versions of the model - Llama-3.1-8B-bnb-4bit and Llama-3.1-8B-Instruct-bnb-4bit - were fine-tuned using parameter-efficient fine-tuning (PEFT) with LoRA adapters and 4-bit quantized models, leveraging the Unsloth framework for accelerated and resource-efficient training. A custom dataset of 6000 legal question-answer pairs was curated from Jordanian laws and formatted into structured prompts. Performance was evaluated using the BLEU and the ROUGE metrics to compare the fine-tuned models to their respective base versions. Results demonstrated improved legal reasoning and accuracy while achieving resource efficiency through quantization and optimized fine-tuning strategies. This work underscores the potential of adapting large language models for Arabic legal domains and highlights effective techniques for fine-tuning domain-specific tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2601_17364
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Parameter Efficient Fine Tuning Llama 3.1 for Answering Arabic Legal Questions: A Case Study on Jordanian Laws
Fasha, Mohammed
Hammo, Bassam
Sowan, Bilal
Barham, Husam
Nsour, Esam
Computation and Language
Artificial Intelligence
This study uses Jordanian law as a case study to explore the fine-tuning of the Llama-3.1 large language model for Arabic question-answering. Two versions of the model - Llama-3.1-8B-bnb-4bit and Llama-3.1-8B-Instruct-bnb-4bit - were fine-tuned using parameter-efficient fine-tuning (PEFT) with LoRA adapters and 4-bit quantized models, leveraging the Unsloth framework for accelerated and resource-efficient training. A custom dataset of 6000 legal question-answer pairs was curated from Jordanian laws and formatted into structured prompts. Performance was evaluated using the BLEU and the ROUGE metrics to compare the fine-tuned models to their respective base versions. Results demonstrated improved legal reasoning and accuracy while achieving resource efficiency through quantization and optimized fine-tuning strategies. This work underscores the potential of adapting large language models for Arabic legal domains and highlights effective techniques for fine-tuning domain-specific tasks.
title Parameter Efficient Fine Tuning Llama 3.1 for Answering Arabic Legal Questions: A Case Study on Jordanian Laws
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2601.17364