解决多语言翻译中反向翻译质量下降问题，提升小模型跨语言能力。
NiuTrans.LMT: Toward Inclusive and Scalable Multilingual Machine Translation with LLMs
- 通过策略性降采样缓解方向性退化，防止模型学走捷径。
- 40亿参数模型性能媲美更大规模基线，覆盖60语言234方向。
- 开源中文-英文核心模型，适合多语言翻译研究与应用。
大语言模型虽推动了多语言机器翻译（MMT）发展，但在扩展至多种语言时，保持各方向翻译质量仍具挑战。本文发现，对多向平行数据采用对称式多语言监督微调（SFT）时，若以某一枢纽语言（如英语）为中心重复使用数据，反向翻译（如X→英语）性能会显著下降，此现象称为方向性退化，源于过度的多对一映射导致模型捷径学习。为此，提出策略性降采样（SD），有效缓解该退化。进一步提出并行多语言提示（PMP），在训练中引入辅助平行句以促进跨语言迁移，并支持测试阶段可选增强。我们构建了大型多语言翻译系统NiuTrans.LMT，包含四种规模（0.6B/1.7B/4B/8B）的模型，覆盖60种语言、234个翻译方向。全面评估表明，该系统在开源多语言翻译系统中具有竞争力，其中40亿参数模型表现不逊于甚至优于更大规模基线。相关模型与资源已公开，以支持包容性与可扩展的多语言翻译。
原文摘要 · Abstract (English)
Large language models have significantly advanced Multilingual Machine Translation (MMT), yet scaling to many languages while keeping quality robust across directions remains challenging. In this paper, we identify a failure mode of multilingual supervised fine-tuning (SFT) on multi-way parallel data: when such data are reused symmetrically around a pivot language (e.g., English), performance on reverse directions (X $\to$ pivot) can drop substantially. We term this phenomenon Directional Degeneration and attribute it to excessive many-to-one mappings, which encourage shortcut learning. We propose Strategic Downsampling (SD), a simple yet effective method to mitigate this degeneration. In addition, we introduce Parallel Multilingual Prompting (PMP), which augments translation instructions with an auxiliary parallel sentence to promote cross-lingual transfer during training and enables optional test-time enhancement when auxiliary translations are available. We further develop \textbf{NiuTrans.LMT} (\textbf{L}arge-scale \textbf{M}ultilingual \textbf{T}ranslation, abbreviated as \textbf{LMT}), a Chinese-English-centric suite of multilingual translation models spanning four sizes (0.6B/1.7B/4B/8B) and covering 60 languages and 234 directions. Comprehensive evaluations show that LMT is competitive among open-source MMT systems, and that our 4B LMT model performs on par with or better than substantially larger baselines. We release our models and project resources to support inclusive and scalable MMT.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。