arXiv：cs.CL· Ruoxi Liu, Philipp Koehn·· 4 小时前AI 评分36
RT-SFT：用往返翻译从非平行语料实现文本风格迁移
RT-SFT: Text Style Transfer from Non-Parallel Corpora by Roundtrip Translation
AI 导读
RT-SFT 利用神经机器翻译的往返翻译剥离风格信号，从单语语料生成伪平行语料，再以 LoRA 微调指令微调 LLM 作为风格化器。在四个风格领域上，该方法大幅超越 few-shot 上下文学习等 SOTA 方法。研究还报告了面向术语和命名规范严格的专家风格领域的有效检索增强方法。
正文
Abstract:Text style transfer (TST) is naturally a supervised task - rewrite a sentence in a target style while preserving its meaning - yet the parallel corpora that supervision requires exist for only a handful of style domains. A common workaround is to *normalize* an input into a style-agnostic intermediate and then *stylize* it into the target style, but the normalizer is typically a lightweight, task-specific paraphraser applied only at test time, feeding a correspondingly small stylizer. We observe that a style-stripping normalizer already exists at scale: neural MT systems trained on hundreds of millions of general-domain sentence pairs preserve content while regressing toward generic phrasing, so roundtrip translation through a pivot language strips stylistic signal without any task-specific training. This turns normalization from an inference-time patch into a data-generation tool. Roundtrip-translating a monolingual in-style corpus yields a pseudo-parallel corpus on which we LoRA-finetune an instruction-tuned LLM as the stylizer (RT-SFT); applying the same normalizer to test queries keeps that stylizer in-distribution. We show that across four style domains, RT-SFT outperforms state-of-the-art methods, such as few-shot in-context learning, by considerable margins. We also report on effective retrieval augmentation methods for expert style domains with strict terminology and naming conventions.
| Comments: | 9 pages, figures, 4 tables |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2602.15013 [cs.CL] |
| (or arXiv:2602.15013v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2602.15013 arXiv-issued DOI via DataCite |
Submission history
From: Ruoxi Liu [view email]
[v1]
Mon, 16 Feb 2026 18:52:43 UTC (308 KB)
[v2]
Thu, 1 Oct 2026 18:42:36 UTC (343 KB)
来源：arXiv：cs.CL · arxiv.org