风格重混：通过风格要素的精馏与扰动实现可解释的作者身份隐匿

摘要

作者身份混淆——即通过重写文本来刻意隐藏作者身份——是一项重要但具有挑战性的任务。当前基于大语言模型的方法缺乏可解释性与可控性，常常忽略作者特有的风格特征，导致整体鲁棒性不足。针对这一问题，我们提出了StyleRemix，一种自适应且可解释的混淆方法，通过对原始文本中特定的细粒度风格元素进行扰动来实现混淆。该方法采用预训练的低秩自适应模块，能够沿不同风格维度（如正式度与文本长度）对输入文本进行定向重写，同时保持较低的计算成本。经自动评估与人工评估验证，StyleRemix在多个领域均优于现有基线模型及参数量更大的语言模型。此外，我们发布了AuthorMix数据集（包含来自14位不同作者、4大领域的3万篇高质量长文本）以及DiSC平行语料库（涵盖7个风格维度、16个独特方向的1500篇文本）。

English

Authorship obfuscation, rewriting a text to intentionally obscure the identity of the author, is an important but challenging task. Current methods using large language models (LLMs) lack interpretability and controllability, often ignoring author-specific stylistic features, resulting in less robust performance overall. To address this, we develop StyleRemix, an adaptive and interpretable obfuscation method that perturbs specific, fine-grained style elements of the original input text. StyleRemix uses pre-trained Low Rank Adaptation (LoRA) modules to rewrite an input specifically along various stylistic axes (e.g., formality and length) while maintaining low computational cost. StyleRemix outperforms state-of-the-art baselines and much larger LLMs in a variety of domains as assessed by both automatic and human evaluation. Additionally, we release AuthorMix, a large set of 30K high-quality, long-form texts from a diverse set of 14 authors and 4 domains, and DiSC, a parallel corpus of 1,500 texts spanning seven style axes in 16 unique directions

风格重混：通过风格要素的精馏与扰动实现可解释的作者身份隐匿

StyleRemix: Interpretable Authorship Obfuscation via Distillation and Perturbation of Style Elements

摘要

Support