大语言模型模仿个体作者的能力有多强？一项对比评估研究
How Well Can Large Language Models Impersonate Individual Authors? A Comparative Evaluation
AI 导读
一项研究评估了 Gemini 3 Pro、Claude Opus、GPT 5.2、GPT 5.2 Pro、DeepSeek 3.2 和 Grok 4.1 Fast 六个大语言模型模仿个体作者写作风格的能力，采用简单与复杂两类提示词和基于余弦距离的评测指标。
正文 · 原文
Abstract:This study investigates the authorship impersonation capacity of six widely used large language models (LLMs): Gemini 3 Pro, Claude Opus, GPT 5.2, GPT 5.2 Pro, DeepSeek 3.2, and Grok 4.1 Fast. The evaluation was conducted using short product review texts selected to ensure high within author consistency in individual writing style. Two types of in context prompt were designed: a simple prompt with minimal instruction and a complex prompt incorporating explicit stylometric guidance for mimicry. Authorship mimicry was evaluated using a cosine distance based distance to training metric, supplemented by visual analyses of embedding distributions. The results indicate that authorship impersonation remained challenging for most LLMs, with Gemini 3 Pro showing the strongest apparent mimicry performance. However, subsequent overlap analysis revealed that Gemini 3 Pro extensively copied and reused sentences and phrases from the training texts, at levels far exceeding those observed among human authors. Follow up factor analyses showed that training set consistency was the most robust predictor of mimicry success across models, whereas author distinctiveness did not generalise as a reliable predictor. Global embedding visualisations further revealed a consistent macro level pattern: for most models, generated texts occupied regions of the embedding space that were largely separated from the corresponding human training clusters. This indicates limited convergence towards target author style and suggests the presence of model specific lexical and phraseological preferences. The findings highlight the importance of distinguishing genuine stylistic imitation from direct textual reuse and have implications for the forensic analysis of textual evidence involving potential LLM enabled impersonation.
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.05771 [cs.CL] |
| (or arXiv:2611.05771v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2611.05771 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Shunichi Ishihara [view email]
[v1]
Wed, 4 Nov 2026 00:22:54 UTC (7,659 KB)
来源：arXiv cs.AI · arxiv.org