研究人员开发了 CoPiT，一个新颖的翻译流程，旨在解决低资源语言的挑战，特别关注蒙古语。该系统利用蒙古语西里尔字母和传统文字之间的数据可用性不平衡，以西里尔字母作为枢轴。CoPiT 在翻译前解决了传统文字中由文字引起的歧义，从而显著提高了准确性和意义传递。该流程在 BLEU 和 COMET 分数上取得了显著的提升，在可比设置下，开源模型达到了或超过了 GPT-4.1 的性能。此外，CoPiT 有助于创建合成平行数据，有助于缓解现实世界低资源场景中的数据稀缺问题。 AI
影响 提高了低资源语言的翻译能力，可能使更广泛的信息获取和交流成为可能。
排序理由 该集群包含一篇详细介绍机器翻译新方法的学术论文。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
arXiv:2607.05849v1 Announce Type: new Abstract: Low-resource languages remain challenging for machine translation, and Mongolian is a representative case. As a digraphic language, Mongolian is written in both Cyrillic and Traditional scripts, which exhibit a severe imbalance in d…
Low-resource languages remain challenging for machine translation, and Mongolian is a representative case. As a digraphic language, Mongolian is written in both Cyrillic and Traditional scripts, which exhibit a severe imbalance in data availability. While the Cyrillic script is r…
Low-resource languages remain challenging for machine translation, and Mongolian is a representative case. As a digraphic language, Mongolian is written in both Cyrillic and Traditional scripts, which exhibit a severe imbalance in data availability. While the Cyrillic script is r…