AI 中文总结
本研究针对阿拉伯语伊斯兰问答场景，基于微调后的google/gemma-4-12B-it模型构建系统，实现了幻觉检测与可信答案选择的两步任务，在公开数据集上取得了优异性能。
AI 中文摘要
大型语言模型在生成伊斯兰问题的流畅回答时，会引入难以识别的事实错误。本文介绍了我们用于\textsc{HalluScoring 2026}任务2.1“伊斯兰幻觉检测与真相查找”的系统，该任务要求完成统一的两步预测：判断大型语言模型生成的阿拉伯语答案是否存在幻觉，并从六个密切相关的候选选项中选出可信答案。我们使用共享任务提供的伊斯兰知识数据集，该数据集包含600个问答实例，其中341个为幻觉答案，259个为非幻觉答案。我们的系统基于微调后的\texttt{google/gemma-4-12B-it}模型，推理过程中采用确定性解码，对生成的输出进行归一化处理以提取幻觉标签和选定选项。该系统在幻觉检测任务中达到了0.928的Macro-F1分数和0.935的标签准确率，在答案选择任务中达到了0.895的选项准确率，综合分数为0.912，在任务的两个阶段均表现出优异性能。较低的选项选择准确率表明，从看似合理的替代选项中区分出可信答案，比检测幻觉回答更具挑战性。
英文摘要
Large language models can generate fluent responses to Islamic questions while introducing factual errors that are difficult to identify. This paper presents our system for \textsc{HalluScoring 2026} Task 2.1, \textit{Islamic Hallucination Detection and Find the Truth}. The task requires a unified two-step prediction: determining whether an Arabic answer generated by an LLM is hallucinated and selecting the verified answer from six closely related candidate options. We use the Islamic knowledge dataset provided by the shared task, which contains 600 question--answer instances, including 341 hallucinated and 259 non-hallucinated answers. Our system is based on the fine-tuned \texttt{google/gemma-4-12B-it} model and uses deterministic decoding during inference. The generated outputs are normalized to extract the hallucination label and the selected option. The system achieves a Macro-F1 score of 0.928 and a label accuracy of 0.935 for hallucination detection, together with an option accuracy of 0.895 for answer selection. These results yield a combined score of 0.912, demonstrating strong performance across both stages of the task. The lower option-selection accuracy indicates that distinguishing the verified answer from plausible alternatives remains more challenging than detecting hallucinated responses.
