Lab17 @ Ahasis Shared Task 2025: Fine-Tuning and Prompting techniques for Sentiment Analysis of Saudi and Darija Dialects

Al Mukhtar Al Hadhrami, Firas Al Mahrouqi, Mohammed Al Shaaili, Hala Mulki


Abstract
In this paper, we describe our contribution in Ahasis shared task: Sentiment analysis on Arabic Dialects in the Hospitality Domain. Through the presented framework, we explored using two learning strategies tailored to a Large Language Model (LLM) and Transformer-based model variants. While few-shot prompting was used with GPT-4o, fine-tuning was adopted once to refine the essential MARBERT model on the Ahasis dataset and then to utilize a MARBERT variant model, SODA-BERT, that was pretrained on an Omani sentiment dataset and later evaluated with the shared task data.
Anthology ID:
2025.ranlp-ahasis.6
Volume:
Proceedings of the Shared Task on Sentiment Analysis for Arabic Dialects
Month:
September
Year:
2025
Address:
Varna, Bulgaria
Editors:
Maram Alharbi, Salmane Chafik, Saad Ezzini, Ruslan Mitkov, Tharindu Ranasinghe, Hansi Hettiarachchi
Venues:
RANLP | WS
SIG:
Publisher:
INCOMA Ltd., Shoumen, Bulgaria
Note:
Pages:
35–39
Language:
URL:
https://aclanthology.org/2025.ranlp-ahasis.6/
DOI:
Bibkey:
Cite (ACL):
Al Mukhtar Al Hadhrami, Firas Al Mahrouqi, Mohammed Al Shaaili, and Hala Mulki. 2025. Lab17 @ Ahasis Shared Task 2025: Fine-Tuning and Prompting techniques for Sentiment Analysis of Saudi and Darija Dialects. In Proceedings of the Shared Task on Sentiment Analysis for Arabic Dialects, pages 35–39, Varna, Bulgaria. INCOMA Ltd., Shoumen, Bulgaria.
Cite (Informal):
Lab17 @ Ahasis Shared Task 2025: Fine-Tuning and Prompting techniques for Sentiment Analysis of Saudi and Darija Dialects (Al Hadhrami et al., RANLP 2025)
Copy Citation:
PDF:
https://aclanthology.org/2025.ranlp-ahasis.6.pdf