# Narrative Text Generation with a Latent Discrete Plan

arXiv: https://arxiv.org/abs/2010.03272 · HTML: https://mlsanity.com/paper/2010.03272/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Oct 2020.

## Built on by (most cited first, 12 of 12)

- [Of Human Criteria and Automatic Metrics: A Benchmark of the Evaluation of Story Generation](https://mlsanity.com/paper/2208.11646/) (2022)
- [Learning Personalized Alignment for Evaluating Open-ended Text Generation](https://mlsanity.com/paper/2310.03304/) (2023)
- [EtriCA: Event-Triggered Context-Aware Story Generation Augmented by Cross Attention](https://mlsanity.com/paper/2210.12463/) (2022)
- [NGEP: A Graph-based Event Planning Framework for Story Generation](https://mlsanity.com/paper/2210.10602/) (2022)
- [Learning to Plan for Language Modeling from Unlabeled Data](https://mlsanity.com/paper/2404.00614/) (2024)
- [Towards Inter-character Relationship-driven Story Generation](https://mlsanity.com/paper/2211.00676/) (2022)
- [A Cross-Attention Augmented Model for Event-Triggered Context-Aware Story Generation](https://mlsanity.com/paper/2311.11271/) (2023)
- [Exploring Story Generation with Multi-task Objectives in Variational Autoencoders](https://mlsanity.com/paper/2111.08133/) (2021)
- [OpeNLGauge: An Explainable Metric for NLG Evaluation with Open-Weights LLMs](https://mlsanity.com/paper/2503.11858/) (2025)
- [Unsupervised Lead Sheet Generation via Semantic Compression](https://mlsanity.com/paper/2310.10772/) (2023)
- [End-to-end Planner Training for Language Modeling](https://mlsanity.com/paper/2410.12492/) (2024)
- [Towards a Linguistic Evaluation of Narratives: A Quantitative Stylistic Framework](https://mlsanity.com/paper/2604.19261/) (2026)

## Abstract (arXiv)

Past work on story generation has demonstrated the usefulness of conditioning on a generation plan to generate coherent stories. However, these approaches have used heuristics or off-the-shelf models to first tag training stories with the desired type of plan, and then train generation models in a supervised fashion. In this paper, we propose a deep latent variable model that first samples a sequence of anchor words, one per sentence in the story, as part of its generative process. During training, our model treats the sequence of anchor words as a latent variable and attempts to induce anchoring sequences that help guide generation in an unsupervised fashion. We conduct experiments with several types of sentence decoder distributions: left-to-right and non-monotonic, with different degrees of restriction. Further, since we use amortized variational inference to train our model, we introduce two corresponding types of inference network for predicting the posterior on anchor words. We conduct human evaluations which demonstrate that the stories produced by our model are rated better in comparison with baselines which do not consider story plans, and are similar or better in quality relative to baselines which use external supervision for plans. Additionally, the proposed model gets favorable scores when evaluated on perplexity, diversity, and control of story via discrete plan.