LLM推理於機器翻譯：基於思考標記的合成數據生成

摘要

大型推理模型（LRMs）通过设计在回答查询之前的自然语言思维过程，为问题解决带来了新的可能性。尽管它们在数学和编码任务中的能力已广为人知，但它们在机器翻译（MT）任务中的影响仍未得到充分探索。在本研究中，我们探讨了在不同资源水平和多种设置下进行跨语言对机器翻译时生成中间标记的益处。我们发现，“思考标记”并不能帮助LRMs更好地执行机器翻译。这一结果推广到了通过蒸馏人类翻译实践启发的链式思维（CoT）进行微调以在翻译前进行推理的模型。具体而言，使用详细说明如何逐步翻译的合成CoT解释对模型进行微调，并未超越标准的输入输出微调。然而，通过结合模块化翻译特定提示策略的输出构建中间标记，确实带来了改进。我们的发现强调，在微调过程中中间标记的贡献高度依赖于其中是否包含翻译尝试。更广泛地说，我们的结果表明，使用教师来精炼目标翻译或扩展平行语料库，比将它们的CoT解释蒸馏到“思考型”机器翻译模型中更具影响力。

English

Large reasoning models (LRMs) have led to new possibilities in terms of problem-solving, through the devising of a natural language thought process prior to answering a query. While their capabilities are well known across mathematics and coding tasks, their impact on the task of machine translation (MT) remains underexplored. In this work, we explore the benefits of the generation of intermediate tokens when performing MT across multiple language pairs of different levels of resourcedness and multiple setups. We find that "thinking tokens" do not help LRMs better perform MT. This result generalizes to models fine-tuned to reason before translating using distilled chain of thought (CoT) inspired by human translators' practices. Specifically, fine-tuning a model with synthetic CoT explanations detailing how to translate step-by-step does not outperform standard input-output fine-tuning. However, constructing the intermediate tokens by combining the outputs of modular translation-specific prompting strategies results in improvements. Our findings underscore that the contribution of intermediate tokens during fine-tuning highly depends on the presence of translation attempts within them. More broadly, our results suggest that using a teacher to refine target translations or to expand parallel corpora is more impactful than distilling their CoT explanations into "thinking" MT models.

LLM推理於機器翻譯：基於思考標記的合成數據生成

LLM Reasoning for Machine Translation: Synthetic Data Generation over Thinking Tokens

摘要

Support