# How Well Do LLMs Imitate Human Writing Style?

arXiv: https://arxiv.org/abs/2509.24930 · HTML: https://mlsanity.com/paper/2509.24930/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 41 references that are in the map.
- First posted on arXiv 29 Sep 2025.

## Builds on (most cited first, 18 of 18)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Qwen Technical Report](https://mlsanity.com/paper/2309.16609/) (2023)
- [Mixtral of Experts](https://mlsanity.com/paper/2401.04088/) (2024)
- [MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers](https://mlsanity.com/paper/2002.10957/) (2020)
- [DetectGPT: Zero-Shot Machine-Generated Text Detection using Probability Curvature](https://mlsanity.com/paper/2301.11305/) (2023)
- [Prompt Programming for Large Language Models: Beyond the Few-Shot Paradigm](https://mlsanity.com/paper/2102.07350/) (2021)
- [A Survey on LLM-Generated Text Detection: Necessity, Methods, and Future Directions](https://mlsanity.com/paper/2310.14724/) (2023)
- [Can Large Language Models Identify Authorship?](https://mlsanity.com/paper/2403.08213/) (2024)
- [Explainable Authorship Verification in Social Media via Attention-based Similarity Learning](https://mlsanity.com/paper/1910.08144/) (2019)
- [Who Wrote it and Why? Prompting Large-Language Models for Authorship Verification](https://mlsanity.com/paper/2310.08123/) (2023)
- [Authorship Attribution through Function Word Adjacency Networks](https://mlsanity.com/paper/1406.4469/) (2014)
- [Syntactic Recurrent Neural Network for Authorship Attribution](https://mlsanity.com/paper/1902.09723/) (2019)
- [Using Prompts to Guide Large Language Models in Imitating a Real Person's Language Style](https://mlsanity.com/paper/2410.03848/) (2024)
- [InstructAV: Instruction Fine-tuning Large Language Models for Authorship Verification](https://mlsanity.com/paper/2407.12882/) (2024)
- [Writer Identification Using Microblogging Texts for Social Media Forensics](https://mlsanity.com/paper/2008.01533/) (2020)
- [Zero-Shot Prompting and Few-Shot Fine-Tuning: Revisiting Document Image Classification Using Large Language Models](https://mlsanity.com/paper/2412.13859/) (2024)

## Built on by (most cited first, 2 of 2)

- [Authorship Impersonation via LLM Prompting does not Evade Authorship Verification Methods](https://mlsanity.com/paper/2603.29454/) (2026)
- [Detecting GPT-Assisted Writing Using Interpretable Stylometric Features](https://mlsanity.com/paper/2609.26687/) (2026)

## Abstract (arXiv)

Large language models (LLMs) can generate fluent text, but their ability to replicate the distinctive style of a specific human author remains unclear. We present a fast, training-free framework for authorship verification and style imitation analysis. The method integrates TF-IDF character n-grams with transformer embeddings and classifies text pairs through empirical distance distributions, eliminating the need for supervised training or threshold tuning. It achieves 97.5% accuracy on academic essays and 94.5% in cross-domain evaluation, while reducing training time by 91.8% and memory usage by 59% relative to parameter-based baselines. Using this framework, we evaluate five LLMs from three separate families (Llama, Qwen, Mixtral) across four prompting strategies - zero-shot, one-shot, few-shot, and text completion. Results show that the prompting strategy has a more substantial influence on style fidelity than model size: few-shot prompting yields up to 23.5x higher style-matching accuracy than zero-shot, and completion prompting reaches 99.9% agreement with the original author's style. Crucially, high-fidelity imitation does not imply human-like unpredictability - human essays average a perplexity of 29.5, whereas matched LLM outputs average only 15.2. These findings demonstrate that stylistic fidelity and statistical detectability are separable, establishing a reproducible basis for future work in authorship modeling, detection, and identity-conditioned generation.