Skip to content
conftrace_
2026 EMNLP EMNLP 2026

PTUK-NLP at the DialectSentEval 2026 Shared Task: MARBERT-Based Sentiment Classification, Missing-Dialect Effects, and Error Analysis

AbstractWe present our system for Subtask 1 (three-class sentiment classification) of the DialectSentEval 2026 shared task, which evaluates five Arabic dialects using the Multi-Dialect-Sent (MDS-3) hotel-review corpus. Our approach fine-tunes MARBERT with a stratified train/validation split and class-weighted cross-entropy loss. On the official test set, our system achieves an F1 score of 0.8175 and an accuracy of 0.8419, ranking 3rd on the official leaderboard. We identify a substantial dialect mismatch in the dataset: the training set contains no Lebanese reviews, while approximately 20% of the test set is Lebanese. Adding additional Lebanese reviews from OCLAR partially reduces this gap, but does not eliminate it. The error analysis also shows that misclassifications are concentrated in the neutral class and in reviews that express mixed feelings, often with high confidence. These findings highlight the challenges of cross-dialect sentiment classification under unseen-dialect conditions.

More from EMNLP 2026

All EMNLP 2026 papers →