How Well Do LLMs Imitate Human Writing Style?

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 $\mathbf{n}$-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, and 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 23.5 x higher style-matching accuracy than zero-shot, and completion prompting repeatedly 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.