alkhder at AlexandriaX-2026: Context-Aware English-to-Dialectal Arabic Dialogue Translation with QLoRA-Fine-Tuned NileChat-3B

Computer Science > Computation and Language arXiv:2610.22525 (cs) [Submitted on 15 Oct 2026] Title: alkhder at AlexandriaX-2026: Context-Aware English-to-Dialectal Arabic Dialogue Translation with QLoRA-Fine-Tuned NileChat-3B Authors: Hasan Alkhder, Maher Pasa, Maria Hanini, Amro Najjar, Ahmet Zengin.

计算机科学 > 计算与语言 arXiv:2610.22525 (cs) [提交于 2026 年 10 月 15 日] 标题：alkhder 在 AlexandriaX-2026 的表现：基于 QLoRA 微调 NileChat-3B 的上下文感知英语到阿拉伯方言对话翻译。作者：Hasan Alkhder, Maher Pasa, Maria Hanini, Amro Najjar, Ahmet Zengin。

Abstract: This paper describes team Alkhder’s submission to Subtask 1 of the AlexandriaX-2026 shared task, which asks systems to translate English dialogue turns into thirteen Arabic dialects. We fine-tune NileChat-3B-Base with QLoRA on the Alexandria parallel dialogue corpus, training one set of adapter weights for all dialects.

摘要：本文介绍了 Alkhder 团队在 AlexandriaX-2026 共享任务子任务 1 中的提交内容，该任务要求系统将英语对话轮次翻译为十三种阿拉伯方言。我们在 Alexandria 平行对话语料库上使用 QLoRA 对 NileChat-3B-Base 进行了微调，并为所有方言训练了一组适配器权重。

Each turn is translated conditioned on the target country and dialect, the domain, the participant roles, the speaker-addressee gender direction, and the preceding dialogue context. At inference, conversations are translated autoregressively and the model is fed its own previous outputs back as context without using any gold target-side test information.

每一轮翻译都基于目标国家和方言、领域、参与者角色、说话者与听话者的性别指向以及之前的对话上下文进行条件化处理。在推理时，对话以自回归方式进行翻译，模型将自身之前的输出作为上下文反馈，而不使用任何黄金目标侧测试信息。

Under the constrained track, using only Alexandria data, it trains in about one hour and translates the private test set in under two hours. The system reaches a country-macro-averaged spBLEU of 20.36 and chrF++ of 36.64 on both tracks.

在受限赛道下，仅使用 Alexandria 数据，模型训练耗时约一小时，并在两小时内完成了私有测试集的翻译。该系统在两个赛道上均达到了 20.36 的国家宏观平均 spBLEU 和 36.64 的 chrF++。