MudawanSn: A Gold-Standard Wolof-Arabic Parallel Corpus for Machine Translation

MudawanSn：用于机器翻译的沃洛夫语-阿拉伯语黄金标准平行语料库

Abstract: We present MudawanSn, a gold-standard resource of 1,271 sentence-aligned pairs manually translated from Wolof into Modern Standard Arabic (MSA). The source texts are drawn from the MasakhaNER corpus and cover politics, society, religion, and sports in Senegalese news discourse. 摘要： 我们介绍了 MudawanSn，这是一个包含 1,271 个句子对齐的黄金标准资源，由人工将沃洛夫语（Wolof）翻译为现代标准阿拉伯语（MSA）。源文本取自 MasakhaNER 语料库，涵盖了塞内加尔新闻话语中的政治、社会、宗教和体育内容。

Although multilingual resources such as FLORES-200 and NTREX include both Wolof and Arabic, no publicly available parallel corpus is specifically designed for the Wolof-Modern Standard Arabic language pair. We describe the corpus construction protocol, sentence alignment procedure, and quality-control workflow. 尽管 FLORES-200 和 NTREX 等多语言资源同时包含了沃洛夫语和阿拉伯语，但目前尚无专门针对沃洛夫语-现代标准阿拉伯语语言对的公开平行语料库。我们详细描述了语料库的构建协议、句子对齐程序以及质量控制工作流程。

We benchmark four machine translation systems spanning three architectural families: NLLB-200 (600M), mT5-base, and two AfriNLLB variants, showing that fine-tuning on MudawanSn yields substantial improvements in both translation directions. 我们对涵盖三种架构系列的四种机器翻译系统进行了基准测试：NLLB-200 (600M)、mT5-base 以及两个 AfriNLLB 变体。结果表明，在 MudawanSn 上进行微调可在两个翻译方向上实现显著的性能提升。

The best-performing model, AfriNLLB-12, achieves 7.76 BLEU and 30.72 chrF++ for Wolof-to-Arabic, and 8.75 BLEU and 33.08 chrF++ for Arabic-to-Wolof. The corpus is released under the CC BY-NC license and is publicly available on Hugging Face and GitHub. 表现最好的模型 AfriNLLB-12 在沃洛夫语到阿拉伯语的翻译中达到了 7.76 的 BLEU 分数和 30.72 的 chrF++ 分数，在阿拉伯语到沃洛夫语的翻译中达到了 8.75 的 BLEU 分数和 33.08 的 chrF++ 分数。该语料库以 CC BY-NC 许可证发布，并已在 Hugging Face 和 GitHub 上公开。