Strategies for Structuring Story Generation

Abstract

Writers generally rely on plans or sketches to write long stories, but most current language models generate word by word from left to right. We explore coarse-to-fine models for creating narrative texts of several hundred words, and introduce new models which decompose stories by abstracting over actions and entities. The model first generates the predicate-argument structure of the text, where different mentions of the same entity are marked with placeholder tokens. It then generates a surface realization of the predicate-argument structure, and finally replaces the entity placeholders with context-sensitive names and references. Human judges prefer the stories from our models to a wide range of previous approaches to hierarchical text generation. Extensive analysis shows that our methods can help improve the diversity and coherence of events and entities in generated stories.

References in corpus (17)

• Teaching Machines to Read and Comprehend
• Building Natural-Language Generation Systems
• Narrative Planning: Balancing Plot and Character
• Non-Autoregressive Neural Machine Translation
• Abstractive Text Summarization Using Sequence-to-Sequence RNNs and Beyond
• Pay Less Attention with Lightweight and Dynamic Convolutions
• Insertion Transformer: Flexible Sequence Generation via Insertion Operations
• Event Representations for Automated Story Generation with Deep Neural Nets
• Trainable Methods for Surface Natural Language Generation
• Visual Storytelling
• Get To The Point: Summarization with Pointer-Generator Networks
• Story Generation from Sequence of Independent Short Descriptions
• Generating Wikipedia by Summarizing Long Sequences
• Insertion-based Decoding with automatically Inferred Generation Order
• A Neural Language Model for Dynamically Representing the Meanings of Unknown Words and Entities in a Discourse
• Dynamic Entity Representations in Neural Language Models
• Building Language Models for Text with Named Entities

Cited by in corpus (10)

• Evaluation of Text Generation: A Survey
• Narrative Interpolation for Generating and Understanding Stories
• Topic-Preserving Synthetic News Generation: An Adversarial Deep Reinforcement Learning Approach
• Induction and Reference of Entities in a Visual Story
• Towards Faithfulness in Open Domain Table-to-text Generation from an Entity-centric View
• Local and Global Context-Based Pairwise Models for Sentence Ordering
• GraphPlan: Story Generation by Planning with Event Graph
• THEaiTRE: Artificial Intelligence to Write a Theatre Play
• Topological Sort for Sentence Ordering
• High Quality Real-Time Structured Debate Generation