Research map / Architectures and efficiency / Attention and transformers
Generating Coherent Narratives by Learning Dynamic and Discrete Entity States with a Contrastive Framework
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 |
|---|---|---|---|
| Use dynamic discrete entity states learned with a contrastive objective | BART-Base narrative generation · Wikiplots test | B-1: 29.81 -> 31.81 | Helped |
| Use dynamic discrete entity states learned with a contrastive objective | BART-Base narrative generation · Wikiplots test | MAUVE: 80.16 -> 87.88 | Helped |
| Use dynamic discrete entity states learned with a contrastive objective | BART-Base narrative generation · CNN News test | B-1: 33.07 -> 33.83 | Helped |
| Use dynamic discrete entity states learned with a contrastive objective | BART-Base narrative generation · CNN News test | MAUVE: 87.25 -> 91.02 | Helped |
| Use dynamic discrete entity states learned with a contrastive objective | BART-Base narrative generation · CNN News test | Rpt-16: 14.37 -> 14.29 | Helped |
| Add entity state attention layer (ablation) | ERIC · Wikiplots test | B-1: 30.12 -> 31.81 | Helped |
| Add entity state attention layer (ablation) | ERIC · Wikiplots test | MAUVE: 83.29 -> 87.88 | Helped |
| Add entity state attention layer (ablation) | ERIC · Wikiplots test | Rpt-16: 16.75 -> 16.22 | Helped |
| Add entity state representations and contrastive learning (ablation) | ERIC · Wikiplots test | B-1: 28.84 -> 31.81 | Helped |
| Add entity state representations and contrastive learning (ablation) | ERIC · Wikiplots test | MAUVE: 83.18 -> 87.88 | Helped |
| Add entity states, contrastive learning, and next-mentioned-entity prediction (ablation) | ERIC · Wikiplots test | B-1: 28.63 -> 31.81 | Helped |
| Add entity states, contrastive learning, and next-mentioned-entity prediction (ablation) | ERIC · Wikiplots test | MAUVE: 73.01 -> 87.88 | Helped |
Closest papers by meaning
- Towards Coherent and Consistent Use of Entities in Narrative Generation (ICML 2022 · similarity 0.89)
- Event Representations for Automated Story Generation with Deep Neural Nets (AAAI 2018 · similarity 0.87)
- The Belief State Transformer (ICLR 2025 · similarity 0.84)
- Parameterized Synthetic Text Generation with SimpleStories (NeurIPS 2025 · similarity 0.84)
- Compressed and Smooth Latent Space for Text Diffusion Modeling (NeurIPS 2025 · similarity 0.84)
- Boosting Consistency in Story Visualization with Rich-Contextual Conditional Diffusion Models (AAAI 2025 · similarity 0.83)
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.