arXiv：cs.CL· Hao Wang, Yangyang Liu, Xiaohu Zhao, Heng Liu, Zifu Shang, Tianhao Li, Ruize Gao, Jialong Tang, Liangying Shao, Haoran Wei, Baosong Yang, Linlong Xu, Longyue Wang, Weihua Luo·· 6 小时前
Qwen-MT-Lumen 亮相 WMT2026：通过质量感知训练与同策略蒸馏实现指令专用翻译
Qwen-MT-Lumen at WMT2026: Instruction-Specialized Translation via Quality-Aware Training and On-Policy Distillation
AI 导读
阿里团队提交 WMT 2026 通用机器翻译共享任务的 Qwen-MT-Lumen（Lumen）采用四阶段流程，通过持续预训练与监督微调、在线偏好优化、同策略蒸馏及结构感知推理，将更大教师模型的翻译指令遵循能力迁移至 Qwen3-14B 学生模型。
正文
Authors:Hao Wang, Yangyang Liu, Xiaohu Zhao, Heng Liu, Zifu Shang, Tianhao Li, Ruize Gao, Jialong Tang, Liangying Shao, Haoran Wei, Baosong Yang, Linlong Xu, Longyue Wang, Weihua Luo
Abstract:We present Lumen, The open-source model is available at, our submission to the WMT 2026 General Machine Translation Shared Task. We focus on instruction-specialized translation, where systems must produce accurate translations while satisfying fine-grained requirements on style, terminology, and structured output. Lumen follows a four-stage pipeline. First, continued pre-training and supervised fine-tuning jointly strengthen multilingual translation competence and the ability to follow translation-specific instructions, such as terminology, style, and format constraints. Second, we extend the originally offline M^{2}PO preference-optimization framework into an online preference-optimization stage: the current policy samples candidate translations, an LLM judge assigns continuous direct-assessment scores, and the resulting multi-pair preferences are used to update the model. Third, on-policy distillation transfers stronger translation-instruction-following behavior from a larger teacher model to the Qwen3-14B student, improving translation-instruction adherence while retaining a 14B primary translation backbone. Finally, structure-aware inference and postprocessing preserve required formats, while selective repair addresses clear structural or decoding failures before submission. We participate in 23 translation directions. Under an absolute 1--5 LLM-judge rubric that evaluates translation-instruction following and translation quality separately, Lumen reaches an overall mean score of 4.60. Despite using a 14B translation backbone, it achieves performance comparable to leading proprietary systems in both translation quality and translation-instruction following.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.19758 [cs.CL] |
| (or arXiv:2610.19758v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.19758 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Hao Wang [view email]
[v1]
Wed, 14 Oct 2026 08:12:10 UTC (777 KB)
来源：arXiv：cs.CL · arxiv.org