arXiv:2507.16530cs.CL2025-07

用大模型实现文本风格迁移、作者识别与验证,三者联动提升可解释性。

Learning Text Styles: A Study on Transfer, Attribution, and Verification

  • 基于参数高效微调和对比解耦,分离文本风格与内容特征。
  • 在多个数据集上实现超过90%的作者归属准确率。
  • 适合需要风格控制、版权溯源或反伪造的研究者使用。

本论文围绕文本风格的计算理解与操控,构建了三大核心方向:(1)文本风格迁移(TST),在保持语义内容不变的前提下改变文本的风格属性(如情感倾向、正式程度);(2)作者归属(AA),通过文本的风格指纹识别其作者身份;(3)作者验证(AV),判断两篇文本是否出自同一作者。针对这些任务中的关键挑战,论文采用大语言模型的参数高效适配方法,实现对风格特征的对比解耦,并通过指令微调提升验证过程的可解释性。

原文摘要 · Abstract (English)

This thesis advances the computational understanding and manipulation of text styles through three interconnected pillars: (1) Text Style Transfer (TST), which alters stylistic properties (e.g., sentiment, formality) while preserving content; (2)Authorship Attribution (AA), identifying the author of a text via stylistic fingerprints; and (3) Authorship Verification (AV), determining whether two texts share the same authorship. We address critical challenges in these areas by leveraging parameter-efficient adaptation of large language models (LLMs), contrastive disentanglement of stylistic features, and instruction-based fine-tuning for explainable verification.

风格迁移作者识别大模型应用

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。