Frankentext: Stitching random text fragments into long-form narratives

TL;DR

Frankentexts is a novel long-form narrative generation method that stitches human-written snippets into coherent stories, improving quality and originality while challenging AI detection and raising authorship questions.

Contribution

The paper introduces Frankentexts, a new paradigm for narrative generation that leverages snippet copying and selection, enhancing diversity and coherence over standard LLM outputs.

Findings

Frankentexts significantly outperform vanilla LLMs in quality, diversity, and originality.

72% of Frankentexts are misclassified as human-written by a state-of-the-art detector.

Human evaluators praise the narratives for inventiveness and vividness, despite some tonal and grammatical issues.

Abstract

We introduce Frankentexts, a long-form narrative generation paradigm that treats an LLM as a composer of existing texts rather than as an author. Given a writing prompt and thousands of randomly sampled human-written snippets, the model is asked to produce a narrative under the extreme constraint that most tokens (e.g., 90%) must be copied verbatim from the provided paragraphs. This task is effectively intractable for humans: selecting and ordering snippets yields a combinatorial search space that an LLM implicitly explores, before minimally editing and stitching together selected fragments into a coherent long-form story. Despite the extreme challenge of the task, we observe through extensive automatic and human evaluation that Frankentexts significantly improve over vanilla LLM generations in terms of writing quality, diversity, and originality while remaining coherent and relevant to the prompt. Furthermore, Frankentexts pose a fundamental challenge to detectors of AI-generated text: 72% of Frankentexts produced by our best Gemini 2.5 Pro configuration are misclassified as human-written by Pangram, a state-of-the-art detector. Human annotators praise Frankentexts for their inventive premises, vivid descriptions, and dry humor; on the other hand, they identify issues with abrupt tonal shifts and uneven grammar across segments, particularly in longer pieces. The emergence of high-quality Frankentexts raises serious questions about authorship and copyright: when humans provide the raw materials and LLMs orchestrate them into new narratives, who truly owns the result?