CLAIDec 20, 2022

DOC: Improving Long Story Coherence With Detailed Outline Control

arXiv:2212.10077v3243 citationsh-index: 85
Originality Incremental advance
AI Analysis

This addresses the challenge of coherent long-form story generation for applications in creative writing and interactive storytelling, representing a strong specific gain rather than a foundational advancement.

The paper tackles the problem of maintaining long-range plot coherence in automatically generated multi-thousand-word stories by introducing the Detailed Outline Control (DOC) framework, which achieved substantial improvements over a strong baseline with absolute gains of 22.5% in plot coherence, 28.2% in outline relevance, and 20.7% in interestingness in human evaluations.

We propose the Detailed Outline Control (DOC) framework for improving long-range plot coherence when automatically generating several-thousand-word-long stories. DOC consists of two complementary components: a detailed outliner and a detailed controller. The detailed outliner creates a more detailed, hierarchically structured outline, shifting creative burden from the main drafting procedure to the planning stage. The detailed controller ensures the more detailed outline is still respected during generation by controlling story passages to align with outline details. In human evaluations of automatically generated stories, DOC substantially outperforms a strong Re3 baseline (Yang et al., 2022) on plot coherence (22.5% absolute gain), outline relevance (28.2%), and interestingness (20.7%). Humans also judged DOC to be much more controllable in an interactive generation setting.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes