arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2310.00436cs.CL
收藏

领域偏移下的作者身份识别:风格度量与学习型作者表征综述

Authorship identification under domain shift: a survey of stylistic measures and learned author representations

Indiana University Bloomington(印第安纳大学伯明顿分校)

Haining Wang

AI总结:

本综述系统梳理领域偏移下作者身份识别的证据,强调主题独立性取决于特征编码、评分与评估划分的组合,并指导风格表征选择与泛化验证。

摘要:

作者身份识别利用写作中的模式来推断文本的作者,但这些模式也反映了主题、体裁和语域。本综述认为,主题独立性并非风格特征本身的属性,而是特征与其编码、评分规则及评估划分共同作用的结果,因此问题在于当领域变化时,哪些组合能保留作者差异。我们将证据组织为四个层面:领域与评估条件、语言度量、表征、以及评分与决策规则,并考察从经典频率度量到学习型作者表征及语言模型介导写作的各类研究内比较。这些比较表明,为何仅凭特征清单无法解释性能:归一化与编码会改变结果,作者监督学习可能保留内容信息,评估协议可能逆转模型排名。因此,成功的识别、风格相似性和泛化需要分别的证据。文献提供了一种实用途径,通过将作者身份与具体写作情境分离的设计来获取证据:同一作者跨体裁写作、不同作者在同一对话内写作,以及内容受控的平行文本。这些设计也有助于评估内容掩蔽和语言模型改写,其效果取决于剩余的可识别信息。我们利用这些证据来指导风格表征的选择以及确定其泛化范围所需的比较。

英文摘要:

Authorship identification uses patterns in writing to infer who wrote a text, but those patterns also reflect topic, genre, and register. This survey argues that topic-independence is not a property of a stylistic feature but of the feature together with its encoding, its scoring rule, and the evaluation split, so the question is which combinations preserve author differences when the domain changes. We organize the evidence in four layers, domains and evaluation conditions, linguistic measures, representations, and scoring and decision rules, and examine within-study comparisons from classical frequency measures to learned author representations and language-model-mediated writing. These comparisons show why feature inventories alone do not explain performance: normalization and encoding change results, author-supervised learning can retain content, and evaluation protocols can reverse model rankings. Successful identification, stylistic similarity, and generalization therefore require separate evidence. The literature offers a practical way to obtain it through designs that separate author identity from particular circumstances of writing: the same authors writing across genres, different authors writing within the same conversation, and parallel texts with controlled content. These designs also help evaluate content masking and language-model rewriting, whose effects depend on what identifying information remains. We use this evidence to guide the choice of stylistic representations and the comparisons needed to establish where they generalize.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑