This paper describes the multi-strategy ensemble approach that has been used to develop the model submitted to the Abductive Event Reasoning shared task. The proposed model combines semantic similarity, causal pattern recognition, and Large Language Models (LLMs) to identify causal relationships between news events and their causes. Our system achieved competitive performance by integrating semantic embedding-based similarity, explicit causal pattern matching, keyword overlap analysis, temporal alignment scoring, and LLM-enhanced reasoning. Our system achieved accuracies of 65.4% and 43.2% on the development set using the LLM-enhanced configuration and the non-LLM ensemble, respectively. The final score using the test set on the leaderboard is 0.3.

In this paper, we present our system, which was submitted to SemEval-2026 Task 9 (Subtask 1: Polarization Detection) and focuses on binary classification of polarized content in Arabic social media text. To address Arabic linguistic variations, we propose a single-model approach that combines fine-tuned AraBERT with synonym-based data augmentation. On the Arabic bind set, our method achieves a competitive macro F1-score of 0.831 and an accuracy of 0.833. Among the 45 participating teams, our system ranked 11th overall, with a performance gap of 0.018 macro F1 from the top-ranked team (0.8488). The results show that a fine-tuned AraBERT with synonym replacement is a strong, simple, and reproducible baseline that outperforms more complex setups in dealing with Arabic attitude polarization nuances.

This paper presents a multistage ensemble-based approach developed for the HalluScoring-2026 shared task on Arabic hallucination detection. The proposed system combines three Arabic pretrained language encoders: MARBERTv2, CAMeLBERT, and ARBERT. The diverse linguistic representations captured by these models enable the ensemble to improve the robustness and discriminative capability of hallucination detection compared to the individual encoders. On the development set, the ensemble achieved an AUC-ROC of 0.9413, Macro-F1 of 0.8704, and an AUC-PR of 0.9610, outperforming the individual models. On the official test set, the system achieved an AUC-ROC of 0.7717, Macro-F1 of 0.6576, and an AUC-PR of 0.9181. These results secured the highest overall ranking among all submitted systems, placing our system first in the HalluScoring-2026 shared task. Further analysis reveals substantial variation in detection performance across generator models and linguistic phenomena, with subtle factual deviations, dialectal and culturally specific expressions, and unusually long responses posing particular challenges. Overall, the findings demonstrate the effectiveness of combining distinct Arabic-specific pretrained encoders for robust Arabic hallucination detection.

We describe REGLAT’s submission to AlexandriaX-2026 Subtask 3, which requires character-offset error spans in dialectal Arabic machine translations and a Linguistically Motivated Multidimensional Quality Metrics (LQM) category for each span. The system separates localization from classification. Four CAMeLBERT-DA and MARBERTv2 BIO taggers project token probabilities to a shared character grid; whitespace-segment probabilities are then thresholded, bridged, and optionally dilated. A second four-member ensemble classifies detected spans using the marked hypothesis, English source, and translation direction, with class weighting and fitted logit biases. REGLAT ranked fourth out of six teams with an official Overall Score of 0.4021 (Overlap F1 0.4272; Error-Class score 0.3770). Its Exact Match F1 of 0.0041 exposes severe boundary imprecision. We document a scorer mismatch in development-time selection and therefore treat earlier development values as proxy-metric results, not official-metric estimates.

The rapid advancement of large language models necessitates robust methods for detecting AI-generated Arabic text. This paper presents our system for distinguishing human-written from machine-generated Arabic content. We propose a weighted ensemble combining AraBERTv2 and BERT-base-arabic, trained via 5-fold stratified cross-validation with class-balanced loss functions. Our methodology incorporates Arabic text normalization, strategic data augmentation using 16,678 samples from external scientific abstracts, and threshold optimization prioritizing recall. On the official test set, our system achieved an F1-score of 0.763, an accuracy of 0.695, a precision of 0.624, and a recall of 0.980, demonstrating strong detection of machine-generated texts with minimal false negatives at the cost of elevated false positives. Analysis reveals critical insights into precision-recall trade-offs and challenges in cross-domain generalization for Arabic AI text detection.

In this paper, we demonstrate the system submitted to the shared task of medical text classification in Arabic. We proposed a single-model approach based on fine-tuned LLM-based embedding combined with hierarchical classical classifiers, achieving a competitive macro F1-score of 0.46 on the blind test set. We explored various modeling strategies, including tree-based ensembles, LLM, and hierarchical correction for rare classes, highlighting the effectiveness of domain-specific fine-tuning in low-resource settings. The results demonstrate that a single fine-tuned Arabic BERT variant can serve as a strong baseline in extreme imbalance scenarios, outperforming more complex ensembles in simplicity and reproducibility.