Research map / Large language models / Language, safety and interpretability
Event Representations for Automated Story Generation with Deep Neural Nets
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| generalize event words using named-entity tags, WordNet, and VerbNet | event2event · Wikipedia movie plots | Perplexity: 748.914 -> 54.231 | Helped |
| generalize event words using named-entity tags, WordNet, and VerbNet | event2event · Wikipedia movie plots | BLEU: 0.1880 -> 0.0575 | Hurt |
| replace person names with numbered NE tags (ablation) | Original Words Baseline · Wikipedia movie plots | Perplexity: 748.914 -> 166.646 | Helped |
| add genre information to generalized events (ablation) | event2event · Wikipedia movie plots | Perplexity: 54.231 -> 48.041 | Helped |
| add genre information to generalized events (ablation) | event2event · Wikipedia movie plots | BLEU: 0.0575 -> 0.0525 | Hurt |
| use two-event input and predict two successor events (ablation) | Generalized event2event · Wikipedia movie plots | BLEU: 0.0575 -> 0.1549 | Helped |
| continue NE numbering across input-output pairs in event bigrams (ablation) | Generalized Bigram · Wikipedia movie plots | BLEU: 0.1549 -> 0.1567 | Helped |
| add genre information to generalized event bigrams (ablation) | Generalized Bigram · Wikipedia movie plots | BLEU: 0.1549 -> 0.1102 | Hurt |
| force multiple events extracted from a sentence into sequence (ablation) | Generalized event2event · Wikipedia movie plots | Perplexity: 54.231 -> 58.562 | Hurt |
| map all events in one sentence to all events in the following sentence (ablation) | Generalized event2event · Wikipedia movie plots | Perplexity: 54.231 -> 45.223 | Helped |
| generalize event and sentence representations rather than using original words | event2sentence · Wikipedia movie plots | BLEU: 0.0016 -> 0.0331 | Helped |
| split and prune sentences while retaining original words (ablation) | event2sentence · Wikipedia movie plots | BLEU: 0.0016 -> 0.0764 | Helped |
Closest papers by meaning
- Story Realization: Expanding Plot Events into Sentences (AAAI 2020 · similarity 0.93)
- What Happens Next? Event Prediction Using a Compositional Neural Network Model (AAAI 2016 · similarity 0.90)
- A Character-Centric Neural Model for Automated Story Generation (AAAI 2020 · similarity 0.89)
- Plan-and-Write: Towards Better Automatic Storytelling (AAAI 2019 · similarity 0.88)
- Intelligent Grimm - Open-ended Visual Storytelling via Latent Diffusion Models (CVPR 2024 · similarity 0.84)
- Is Your World Simulator a Good Story Presenter? A Consecutive Events-Based Benchmark for Future Long Video Generation (CVPR 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.