arXiv cs.CL· Aixiu An, Michael Jungo, Eloi Eynard, Mark Drenhaus, Andreas Fischer, Jean Hennebert, S\'ebastien Rumley·· 11 小时前AI 评分31
推理先于翻译：用结构化推理增强法律机器翻译
Reasoning Before Translation: Enhancing Legal Machine Translation with Structured Reasoning
AI 导读
研究对比了监督微调、强化学习与推理模型在法律机器翻译上的效果，以瑞士多语言法律体系为测试场景。经可验证奖励强化学习（RLVR）增强的 Qwen3.5 4B、Qwen3.5 9B 和 Gemma 3 12B 小模型翻译质量大幅提升，并超过监督微调，接近但仍逊于前沿推理模型。重训练范式的收益随模型规模增大而递减，代码与模型已公开。
正文 · 原文
Abstract:Neural machine translation (NMT) in the legal domain is a linguistically and conceptually demanding task, primarily due to the complexity of legal language and the high level of precision it requires. The recent emergence of reasoning-capable language models opens new possibilities for tackling such challenges. They add to a set of other previously proposed techniques to enhance the translation quality, which includes supervised fine-tuning and reinforcement learning.
In this work, we perform a comparison between these various approaches. More particularly, we evaluate small language models such as Qwen3.5 4B, Qwen3.5 9B, and Gemma 3 12B enhanced with various re-training paradigms and compare their performances against frontier reasoning models. We focus on the Swiss legal system, which -- with its unique multilingual statutes -- offers a particularly challenging testbed for reasoning-augmented models. Our results show that the quality of small ``base'' models can be greatly enhanced, and that reinforcement learning with verifiable rewards can be applied to NMT in the legal domain and surpasses the translation quality of supervised fine-tuning. The performance of enhanced small models is close to the one of state-of-the-art reasoning models yet remains inferior. We also note that re-training paradigms yield diminishing returns as model size increase. The code and models are publicly available at this https URL.
| Comments: | Code available at this https URL |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2607.19181 [cs.CL] |
| (or arXiv:2607.19181v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2607.19181 arXiv-issued DOI via DataCite |
Submission history
From: Michael Jungo [view email]
[v1]
Tue, 21 Jul 2026 15:15:56 UTC (45 KB)
[v2]
Fri, 6 Nov 2026 12:50:28 UTC (45 KB)
来源：arXiv cs.CL · arxiv.org