Skip to content
conftrace_
2026 EMNLP EMNLP 2026

NU-SentEval at DialectSentEval Shared Task Subtask-1: Stacking Pre-trained Language Models with Focal Loss for Arabic Dialect Sentiment Analysis

AbstractThis paper presents the system developed by the NU-SentEval team for Subtask 1 (Sentiment Analysis in Dialectal Arabic) of the DialectSentEval 2026 Shared Task. Arabic dialect sentiment classification is challenging due to morphological complexity, orthographic variation across regional dialects, and class imbalances in user reviews. To address these challenges, we propose an ensembled architecture leveraging five diverse Pre-trained Language Models (PLMs): MARBERTv2, MARBERT, CAMeLBERT-DA, AraBERTv0.2-Twitter, and Twitter-XLM-RoBERTa. All models were fine-tuned using a customized Focal Loss objective combined with Label Smoothing to mitigate majority class bias and prevent overconfidence. Out-of-fold probability representations from individual models were aggregated through a Logistic Regression stacking meta-classifier (stacking_c1). On the official test set, our system achieved a Macro F1-score of 0.8141 and an Accuracy of 0.8343, demonstrating the effectiveness of combining specialized dialectal representations with loss regularization and meta-learning.

More from EMNLP 2026

All EMNLP 2026 papers →