用少量样本快速迁移作者风格，保持原意更准确。
AuthorMix: Modular Authorship Style Transfer via Layer-wise Adapter Mixing
- 按层混合独立训练的适配器，实现风格灵活迁移。
- 仅需少量样例即可在低资源目标上超越SOTA。
- 适合需要精准保留语义的文本改写场景。
作者风格迁移旨在改写文本以匹配目标作者的风格，同时保持原意。现有方法通常在大规模语料上训练单一模型以统一建模所有风格，成本高且灵活性差，常牺牲语义保真度换取风格迁移效果。本文提出AuthorMix：一种轻量、模块化且可解释的风格迁移框架。我们在少量高资源作者的小规模数据上分别训练特定风格的LoRA适配器，通过学习到的层级适配器混合机制，仅需少量目标风格样本即可快速构建针对新目标的专用适配模型。AuthorMix在低资源目标上表现优于现有SOTA基线及GPT-5.1，整体得分最高，并在自动与人工评估中显著提升语义保真度。
原文摘要 · Abstract (English)
The task of authorship style transfer involves rewriting text in the style of a target author while preserving the meaning of the original text. Existing style transfer methods train a single model on large corpora to model all target styles at once: this high-cost approach offers limited flexibility for target-specific adaptation, and often sacrifices meaning preservation for style transfer. In this paper, we propose AuthorMix: a lightweight, modular, and interpretable style transfer framework. We train individual, style-specific LoRA adapters on a small set of high-resource authors, allowing the rapid training of specialized adaptation models for each new target via learned, layer-wise adapter mixing, using only a handful of target-style training examples. AuthorMix outperforms existing, SoTA style-transfer baselines-as well as GPT-5.1-for low-resource targets, achieving the highest overall score and substantially improving meaning preservation in both automatic and human evaluations.
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。