arXiv cs.CL· Burte Bayarsaikhan, Serynn Kim, Buru Chang·· 12 小时前AI 评分20
CoPiT：面向传统蒙古文双文种低资源场景的认知枢轴翻译
CoPiT: Cognitive Pivot Translation for Digraphic Low-Resource Mongolian in the Traditional Script
AI 导读
CoPiT 提出一种认知动机驱动的枢轴翻译流程，将传统蒙古文翻译先经由资源较丰富的西里尔蒙古文中转，并在翻译前显式消解传统文种的拼写歧义。在多个骨干模型和目标语言上，CoPiT 普遍优于直接翻译，BLEU 绝对提升显著，COMET 最高提升 1.5-1.6 倍，使强开源模型在可比评测下追平或超过 GPT-4.1。
正文 · 原文
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 data availability. While the Cyrillic script is relatively well-resourced, the Traditional script remains extremely data-scarce and orthographically ambiguous, leading to substantial performance degradation in direct translation. We propose CoPiT, a cognitively motivated pivot-based translation pipeline that exploits this internal resource hierarchy by routing translation through the Cyrillic script. The pipeline explicitly resolves script-induced ambiguity in the Traditional script before translation, enabling more stable and accurate meaning transfer. Across multiple backbone models and target languages, CoPiT generally improves over direct translation, with substantial absolute BLEU improvements and up to 1.5-1.6x gains in COMET. These gains allow strong open-source models to match or outperform GPT-4.1 under comparable evaluation settings. Beyond inference-time improvements, CoPiT enables the construction of synthetic parallel data directly from Traditional-script text, mitigating data scarcity in realistic low-resource scenarios. We release a new multi-script parallel dataset covering Mongolian in both scripts alongside English, Korean, and Russian.
| Comments: | Accepted to EMNLP 2026 (Main Conference) |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2607.05849 [cs.CL] |
| (or arXiv:2607.05849v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2607.05849 arXiv-issued DOI via DataCite |
Submission history
From: Burte Bayarsaikhan [view email]
[v1]
Tue, 7 Jul 2026 05:12:13 UTC (2,881 KB)
[v2]
Mon, 2 Nov 2026 02:22:29 UTC (2,744 KB)
来源：arXiv cs.CL · arxiv.org