Enhancing Arabic Multi-Task Sentiment Analysis Through Distillation and Adversarial Training
Machine Learning and Knowledge Extraction · 2026-04-13 · 8 (4) · 100
DOI: 10.3390/make8040100 ↗At a glance
Journal article in Machine Learning and Knowledge Extraction, published online on 2026-04-13. Written by Hafida Hidani and 2 co-authors. 37 references. The full text is freely available (open access, CC BY 4.0).
Abstract (original language)
The rapid growth of Arabic social media content requires the development of accurate and efficient methods for sentiment analysis. We propose a resource-efficient multi-task learning (MTL) framework for modern standard Arabic (MSA). The model uses a shared AraBERT encoder to jointly predict emotion, polarity, and intention. We integrate knowledge distillation (KD) from a large teacher model, self-distillation (SD) using model self-ensembling, and adversarial training (AT) as a regularization strategy. Experiments conducted on an annotated corpus of MSA tweets demonstrate that all distilled models outperform a fine-tuned multi-task baseline, and the combined KD+SD+AT configuration achieves competitive results. For instance, KD alone raised Macro F1 for emotion from 0.83 to 0.88 and for intention from 0.67 to 0.72. KD+SD+AT achieved the best intention F1 (0.76) and the highest polarity F1 (0.90). Notably, F1-scores for several minority classes show consistent improvement, particularly under KD and combined configurations. Paired t-tests confirm that several improvements, especially those obtained with KD and KD+SD+AT, are statistically significant (p
Cite
APA
Hidani, H., Ouahabi, S. E., & Bouami, M. F. (2026). Enhancing Arabic Multi-Task Sentiment Analysis Through Distillation and Adversarial Training. Machine Learning and Knowledge Extraction, 8(4), 100. https://doi.org/10.3390/make8040100
Vancouver
Hidani H, Ouahabi SE, Bouami MF. Enhancing Arabic Multi-Task Sentiment Analysis Through Distillation and Adversarial Training. Machine Learning and Knowledge Extraction. 2026;8(4):100. doi:10.3390/make8040100
BibTeX
@article{Hidani2026,
author = {Hidani, Hafida and Ouahabi, Safâa El and Bouami, Mouncef Filali},
title = {Enhancing Arabic Multi-Task Sentiment Analysis Through Distillation and Adversarial Training},
journal = {Machine Learning and Knowledge Extraction},
year = {2026},
volume = {8},
number = {4},
pages = {100},
doi = {10.3390/make8040100}
}
EUVIMED