How Well Do LLMs Imitate Human Writing Style?

• URL: http://arxiv.org/abs/2509.24930v1
• Date: Mon, 29 Sep 2025 15:34:40 GMT
• Title: How Well Do LLMs Imitate Human Writing Style?
• Authors: Rebira Jemama, Rajesh Kumar,
• Abstract summary: Large language models (LLMs) can generate fluent text, but their ability to replicate the distinctive style of a specific human author remains unclear.<n>We present a fast, training-free framework for authorship verification and style imitation analysis.<n>It achieves 97.5% accuracy on academic essays and 94.5% in cross-domain evaluation.
• Score: 2.3754840025365183
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: 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.

Related papers

• StyleDecipher: Robust and Explainable Detection of LLM-Generated Texts with Stylistic Analysis [18.44456241158174]
  StyleDecipher is a robust and explainable detection framework.<n>It revisits text detection using combined feature extractors to quantify stylistic differences.<n>It consistently achieves state-of-the-art in-domain accuracy.
  arXiv  Detail & Related papers  (2025-10-14T15:07:27Z)
• Catch Me If You Can? Not Yet: LLMs Still Struggle to Imitate the Implicit Writing Styles of Everyday Authors [9.921537507947473]
  This work presents a comprehensive evaluation of large language models' ability to mimic personal writing styles.<n>We introduce an ensemble of complementary metrics-including authorship attribution, authorship verification, style matching, and AI detection-to robustly assess style imitation.<n>Results show that while LLMs can approximate user styles in structured formats like news and email, they struggle with nuanced, informal writing in blogs and forums.
  arXiv  Detail & Related papers  (2025-09-18T02:18:49Z)
• SCOPE: A Self-supervised Framework for Improving Faithfulness in Conditional Text Generation [55.61004653386632]
  Large Language Models (LLMs) often produce hallucinations, i.e., information that is unfaithful or not grounded in the input context.<n>This paper introduces a novel self-supervised method for generating a training set of unfaithful samples.<n>We then refine the model using a training process that encourages the generation of grounded outputs over unfaithful ones.
  arXiv  Detail & Related papers  (2025-02-19T12:31:58Z)
• Boosting Semi-Supervised Scene Text Recognition via Viewing and Summarizing [71.29488677105127]
  Existing scene text recognition (STR) methods struggle to recognize challenging texts, especially for artistic and severely distorted characters. We propose a contrastive learning-based STR framework by leveraging synthetic and real unlabeled data without any human cost. Our method achieves SOTA performance (94.7% and 70.9% average accuracy on common benchmarks and Union14M-Benchmark.
  arXiv  Detail & Related papers  (2024-11-23T15:24:47Z)
• A Bayesian Approach to Harnessing the Power of LLMs in Authorship Attribution [57.309390098903]
  Authorship attribution aims to identify the origin or author of a document. Large Language Models (LLMs) with their deep reasoning capabilities and ability to maintain long-range textual associations offer a promising alternative. Our results on the IMDb and blog datasets show an impressive 85% accuracy in one-shot authorship classification across ten authors.
  arXiv  Detail & Related papers  (2024-10-29T04:14:23Z)
• Better Zero-Shot Reasoning with Role-Play Prompting [10.90357246745529]
  Role-play prompting consistently surpasses the standard zero-shot approach across most datasets. This highlights its potential to augment the reasoning capabilities of large language models.
  arXiv  Detail & Related papers  (2023-08-15T11:08:30Z)
• Text Classification via Large Language Models [63.1874290788797]
  We introduce Clue And Reasoning Prompting (CARP) to address complex linguistic phenomena involved in text classification. Remarkably, CARP yields new SOTA performances on 4 out of 5 widely-used text-classification benchmarks. More importantly, we find that CARP delivers impressive abilities on low-resource and domain-adaptation setups.
  arXiv  Detail & Related papers  (2023-05-15T06:24:45Z)
• PART: Pre-trained Authorship Representation Transformer [52.623051272843426]
  Authors writing documents imprint identifying information within their texts.<n>Previous works use hand-crafted features or classification tasks to train their authorship models.<n>We propose a contrastively trained model fit to learn textbfauthorship embeddings instead of semantics.
  arXiv  Detail & Related papers  (2022-09-30T11:08:39Z)
• Revisiting Self-Training for Few-Shot Learning of Language Model [61.173976954360334]
  Unlabeled data carry rich task-relevant information, they are proven useful for few-shot learning of language model. In this work, we revisit the self-training technique for language model fine-tuning and present a state-of-the-art prompt-based few-shot learner, SFLM.
  arXiv  Detail & Related papers  (2021-10-04T08:51:36Z)
• MOCHA: A Dataset for Training and Evaluating Generative Reading Comprehension Metrics [55.85042753772513]
  We introduce a benchmark for training and evaluating generative reading comprehension metrics: MOdeling Correctness with Human. s. Using MOCHA, we train a Learned Evaluation metric for Reading Pearson, LERC, to mimic human judgement scores. LERC outperforms baseline metrics by 10 to 36 absolute points on held-out annotations. When we evaluate on minimal pairs, LERC achieves 80% accuracy, outperforming baselines by 14 to 26 absolute percentage points while leaving significant room for improvement.
  arXiv  Detail & Related papers  (2020-10-07T20:22:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.