Get Started
Research questionHow does LLM-generated fiction construct narrative settings differently from human-authored fiction across languages?LLM-generated stories may organize setting and narrative space differently from human fiction, making it difficult to determine whether scenes are grounded in character-environment action or primarily convey atmosphere and perception. These patterns may vary across languages and models.
AI
Natural Language Processing
Research Paper
Latest papersRecent research connected to this question, newest first.How LLMs Build Fictional Worlds: Setting and Narrative Space in AI-Generated Creative StorytellingThe analysis compares 1,000 stories per model in English and German from GPT-4.1, Llama 3.3, Mistral 3.2, and Gemma 3 with human-authored fiction from Project Gutenberg. Setting is categorized as action, perceived, visual, descriptive, or no space using fine-tuned BERT classifiers for each language; the evidence reports model-specific and language-sensitive differences, including greater perceived-space use by LLMs and greater action-space use in the human-authored baseline.research paper · Sep 2, 2026
Related questions
How can LLMs generate diverse stories rather than converge on generic narratives?How can LLMs generate public arguments without collapsing debate into repetitive content and structure?How can authorship attribution remain reliable for long-form LLM text across languages and distribution shifts?How can LLMs synthesize executable machines and improve their designs through physical simulation feedback?
Home
Topics
Search
Library