REGLAT at ArGuard Shared Tasks: An Ensemble-Based Model Integrating NER for Harmful Content Detection in Arabic

arXiv cs.CL Papers
harmful-content-detection arabic-nlp named-entity-recognition text-classification shared-task ensemble-methods

Summary

该论文提出一种两阶段方法用于ArGuard 2026共享任务中的阿拉伯语有害内容检测：先用GLiNER提取有害相关实体，再将实体与原始文本输入微调后的MARBERT进行分类，在开发集上取得0.886的macro-F1分数。

arXiv:2610.37265v1 Announce Type: new Abstract: Detecting harmful content in Arabic is challenging because of dialectal variations and the semantic overlap between different harm categories. This paper presents a two-stage approach for classifying Arabic user prompts into seven categories of harmful content. First, GLiNER is used to extract harm-related entities from each prompt. The extracted entities are then added to the original text and passed to a fine-tuned MARBERT classifier for final classification. The approach combines explicit entity information with the contextual understanding of a pretrained Arabic language model. The system was evaluated on the development set of the ArGuard 2026 shared task. The results achieved a macro-F1 score of 0.886 and an accuracy of 0.897, demonstrating promising performance in fine-grained harmful content detection. The system reported an accuracy of 0.793 and Macro-F score of 0.735 on a blind test set for subtask 1. For subtask 2, it achieved an accuracy of 0.770 and Macro-F score of 0.754. The analysis also showed that distinguishing between related categories remains challenging. These findings suggest that integrating NER with Arabic language models is a promising approach to harmful content detection.
Original Article
View Cached Full Text

Cached at: 10/28/26, 08:19 AM 

# REGLAT at ArGuard Shared Tasks: An Ensemble-Based Model Integrating NER for Harmful Content Detection in Arabic
Source: [https://arxiv.org/abs/2610.37265](https://arxiv.org/abs/2610.37265)
[View PDF](https://arxiv.org/pdf/2610.37265)[HTML \(experimental\)](https://arxiv.org/html/2610.37265v1)

> Abstract:Detecting harmful content in Arabic is challenging because of dialectal variations and the semantic overlap between different harm categories\. This paper presents a two\-stage approach for classifying Arabic user prompts into seven categories of harmful content\. First, GLiNER is used to extract harm\-related entities from each prompt\. The extracted entities are then added to the original text and passed to a fine\-tuned MARBERT classifier for final classification\. The approach combines explicit entity information with the contextual understanding of a pretrained Arabic language model\. The system was evaluated on the development set of the ArGuard 2026 shared task\. The results achieved a macro\-F1 score of 0\.886 and an accuracy of 0\.897, demonstrating promising performance in fine\-grained harmful content detection\. The system reported an accuracy of 0\.793 and Macro\-F score of 0\.735 on a blind test set for subtask 1\. For subtask 2, it achieved an accuracy of 0\.770 and Macro\-F score of 0\.754\. The analysis also showed that distinguishing between related categories remains challenging\. These findings suggest that integrating NER with Arabic language models is a promising approach to harmful content detection\.

## Submission history

From: Hamada Nayel \[[view email](https://arxiv.org/show-email/72ddab2a/2610.37265)\] **\[v1\]**Tue, 27 Oct 2026 13:59:11 UTC \(1,917 KB\)

Similar Articles