Get Started
Research questionHow can long-form story generation preserve narrative consistency as scenes accumulate?As stories grow, generated scenes can introduce temporal, factual, character, commonsense, and stylistic contradictions that propagate into later scenes. The difficulty is maintaining a coherent evolving narrative rather than correcting isolated errors after generation.
AI
Evaluation & Benchmarks
Natural Language Processing
Latest papersRecent research connected to this question, newest first.ConWriter: Transition-Constrained Stateful Long-Form Story Generation with Lightweight Neuro-Symbolic Consistency ControlThe source evaluates training-free long-form story generation on ConStory-Bench across four story-generation tasks, three target lengths, and multiple base language models using the official protocol. It reports narrative-consistency comparisons with direct generation and the DOME baseline.research paper · Sep 4, 2026
Related questions
How can text-to-image models preserve variation in unspecified visual factors under long, semantically dense prompts?How can language models compress conversational and document context without reconstruction while preserving task-relevant information?How can language models sparsely access a growing internal state without limiting it to fixed-size memory?How can long-horizon interactive video world-model training remain reproducible across heterogeneous datasets and incompatible backbones?
Home
Topics
Search
Library