Novels generated by language models show compressed formal variation

• URL: http://arxiv.org/abs/2608.12630v1
• Date: Wed, 12 Aug 2026 22:32:39 GMT
• Title: Novels generated by language models show compressed formal variation
• Abstract summary: This study asks whether repeated AI generation can produce the same range of diversity which is found across human corpora.<n>The research includes MATTR-500, Shannon entropy, average sentence length, readability, and punctuation rate measurements.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: While large language models can generate entire novels, there is little information about the level of formal variation in their output over many generations. Rather than asking whether individual passages can be identified as AI-generated, this study asks whether repeated AI generation can produce the same range of diversity which is found across human corpora. This paper contrasts six corpora based on generation source and target style: twenty novels generated using GPT-5.5 Thinking in a nineteenth-century British realist style, twenty novels generated using Qwen3-14B in a nineteenth-century British realist style, twenty novels generated using each of these models in a contemporary zero style, 205 nineteenth-century human-written British novels, and sixty-five contemporary human-written Zero-Style novels. At the document level, the research includes MATTR-500, Shannon entropy, average sentence length, readability, and punctuation rate measurements. The most robust and reliable result is compression of sentence structure. Repeated generations produce novels that vary far less from one another in sentence structure than human novels do. Compression is also present in the measures of readability, punctuation, and sentence length variability within novels. Lexical measures tend to be similarly compressed, with the exception of Qwen Zero-Style MATTR. Despite having distinct mean stylistic profiles, GPT and Qwen lack a stable pattern of cross-measure correlation. This article therefore distinguishes between variance overclosure, which represents a limited formal range between novels, and a more specific phenomenon of correlational overclosure. This means that an individual AI-generated novel may resemble human fiction stylistically, while a collection of AI-generated novels occupies a much narrower formal range.

Related papers

• StoryScope: Investigating idiosyncrasies in AI fiction [26.913725703912572]
  We find that AI-generated stories cluster in a shared region of narrative space, while human-authored stories exhibit greater diversity.<n>More broadly, these results suggest that differences in underlying narrative construction, not just writing style, can be used to separate human-written original works from AI-generated fiction.
  arXiv  Detail & Related papers  (2026-04-03T15:56:38Z)
• Semantic Novelty Trajectories in 80,000 Books: A Cross-Corpus Embedding Analysis [0.0]
  I analyze semantic novelty trajectories in more than 80,000 books spanning two centuries of English-language publishing.<n> mean paragraph-level novelty is roughly 10% higher in modern books.<n> convergent narrative curves, in which novelty declines toward a settled semantic register, are 2.3x more common in pre-1920 literature.
  arXiv  Detail & Related papers  (2026-03-02T12:20:24Z)
• Generative AI & Fictionality: How Novels Power Large Language Models [1.4847369589597454]
  We study how novels shape the outputs of generative AI.<n>We find that novels leverage familiar attributes and affordances of fiction, while also fomenting new qualities and forms of social response.<n>We argue that if contemporary culture is increasingly shaped by generative AI and machine learning, any analysis of today's various modes of cultural production must account for a relatively novel dimension: computational training data.
  arXiv  Detail & Related papers  (2026-03-01T18:34:02Z)
• ChatGPT-generated texts show authorship traits that identify them as non-human [0.6741942263052466]
  This work examines whether a language model can also be linked to a specific fingerprint.<n>We find that the model can successfully adapt its style depending on whether it is prompted to produce a Wikipedia entry vs. a college essay.<n>Our results suggest that the model prefers nouns to verbs, thus showing a distinct linguistic backbone from humans.
  arXiv  Detail & Related papers  (2025-08-22T13:38:58Z)
• Pastiche Novel Generation Creating: Fan Fiction You Love in Your Favorite Author's Style [48.08453439104245]
  We propose WriterAgent, a novel generation system designed to master the core aspects of literary pastiche.<n>WriterAgent is trained through a curriculum learning paradigm, progressing from low-level stylistic mastery to high-level narrative coherence.<n>We evaluate WriterAgent on multilingual classics like Harry Potter and Dream of the Red Chamber, demonstrating its superiority over baselines.
  arXiv  Detail & Related papers  (2025-02-21T17:40:42Z)
• Unveiling Temporal Trends in 19th Century Literature: An Information Retrieval Approach [5.804963603084041]
  In English literature, the 19th century witnessed a significant transition in styles, themes, and genres.<n>This paper explores the evolution of term usage in 19th century English novels through the lens of information retrieval.
  arXiv  Detail & Related papers  (2025-01-12T15:00:10Z)
• Understanding writing style in social media with a supervised contrastively pre-trained transformer [57.48690310135374]
  Online Social Networks serve as fertile ground for harmful behavior, ranging from hate speech to the dissemination of disinformation. We introduce the Style Transformer for Authorship Representations (STAR), trained on a large corpus derived from public sources of 4.5 x 106 authored texts. Using a support base of 8 documents of 512 tokens, we can discern authors from sets of up to 1616 authors with at least 80% accuracy.
  arXiv  Detail & Related papers  (2023-10-17T09:01:17Z)
• The Stable Entropy Hypothesis and Entropy-Aware Decoding: An Analysis and Algorithm for Robust Natural Language Generation [59.7381286976957]
  We show that human-like'' generations usually lie in a narrow and nearly flat entropy band. We propose an entropy-aware decoding algorithm that respects these entropy bounds.
  arXiv  Detail & Related papers  (2023-02-14T02:02:33Z)
• Twist Decoding: Diverse Generators Guide Each Other [116.20780037268801]
  We introduce Twist decoding, a simple and general inference algorithm that generates text while benefiting from diverse models. Our method does not assume the vocabulary, tokenization or even generation order is shared.
  arXiv  Detail & Related papers  (2022-05-19T01:27:53Z)
• How much do language models copy from their training data? Evaluating linguistic novelty in text generation using RAVEN [63.79300884115027]
  Current language models can generate high-quality text. Are they simply copying text they have seen before, or have they learned generalizable linguistic abstractions? We introduce RAVEN, a suite of analyses for assessing the novelty of generated text.
  arXiv  Detail & Related papers  (2021-11-18T04:07:09Z)
• MixPoet: Diverse Poetry Generation via Learning Controllable Mixed Latent Space [79.70053419040902]
  We propose MixPoet, a novel model that absorbs multiple factors to create various styles and promote diversity. Based on a semi-supervised variational autoencoder, our model disentangles the latent space into some subspaces, with each conditioned on one influence factor by adversarial training. Experiment results on Chinese poetry demonstrate that MixPoet improves both diversity and quality against three state-of-the-art models.
  arXiv  Detail & Related papers  (2020-03-13T03:31:29Z)

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.