Cross-lingual Few-shot Learning for Persian Sentiment Analysis with Incremental Adaptation

Ziaeddin Beheshtifard D epartmen t of Computer E ngineering Islamic Azad University, South Tehran Branch Tehran, Iran zia.beheshti@iau.ac.ir Abstract -- This research examines cross - lingual sentiment analysis using few - shot learning and incremental learning methods in Persian . The main objective is to develop a model capable of performing sentiment analysis in Persian using limited data, while getting prior knowledge from high - resource languages. To achieve this, th re e pre - trained multilingual models ( XLM - RoBERTa, mDeBERTa, and DistilBERT) were employed, which were fine - tuned using few - shot and incremental learning approaches on small samples of Persian dat a from diverse sources, including X, Instagram, Digikala, Snappfood, and Taaghche . This variety enabled the models to learn from a broad range of contexts . Experimental results show that the mDeBERTa and XLM - RoBERTa achieved high performance s, reaching 96% accuracy on Persian sentiment analysis. These findings highlight the effectiveness of combining few - shot learning and incremental learning with multilingual pre - trained models . Sentiment analysis aims to detect and classify emotions expressed in text automatically .