AI summaryⓘ
The authors propose LiteraryBigFive, a method that represents an author's writing style as points along five clear and interpretable axes inspired by personality psychology. Instead of treating writing traits as separate labels, their approach places authors and texts in a shared five-dimensional space based on stylistic contrasts with neutral writing. This allows for better understanding and control of text generation to match specific author styles without needing lots of data or complex tuning. Their experiments show that this method enhances how well generated text reflects an author's style while keeping the meaning intact. The authors also find that their system's style scores align well with known literary opinions, making the model's behavior easy to explain.
Personalized text generationAuthor modelingBig Five personality modelStylistic dimensionsText generation steeringActivation-space contrastsLiterary style analysisSemantic fidelityInterpretabilityCreative writing support
Authors
Jinghui Zhang, Lang Gao, Ao Li, Mingzhe Li, Ruihong Zeng, Zirui Song, Kentaro Inui, Xiuying Chen
Abstract
Personalized text generation for authors and literary writing is essential for applications such as adaptive writing assistants, creative support tools, and computational literary analysis. However, existing approaches to author modeling and personalization often represent writing behavior as independent labels, requiring large-scale corpus collection or fine-tuning for each author or stylistic category. Such formulations are costly, difficult to interpret, and poorly suited for generalizing across authors. Inspired by the Big Five model's dimensional view of personality, we propose LiteraryBigFive, a framework that reframes authorial writing characteristics as coordinates within a unified and interpretable space. In this space, we derive each interpretable axis (e.g., Classicism, Emotionality) from activation-space contrasts between author-written and neutral passages, yielding distinct stylistic dimensions that allow texts or authors to be positioned within a five-dimensional system. Beyond localizing different authors, we further introduce an interpretable steering mechanism, which adaptively guides text generation toward target coordinates to perform author-personalized writing. Experimental results show that LiteraryBigFive improves authorial expressiveness while preserving semantic fidelity. The derived author per-axis scores strongly correlate with real-world literary consensus, offering transparent and interpretable explanations of author-specific generation behavior: https://github.com/Znull-1220/LiteraryBigFive.