PlotTwist: A Creative Plot Generation Framework with Small Language Models

2026-03-17

AI Summary

This work addresses the challenge of generating high-quality narratives—characterized by coherent structure, vivid characters, and emotional depth—with small language models (≤5B parameters) under resource constraints. The authors propose a structured story generation framework that decomposes the process into three stages: evaluation, generation, and validation. This framework integrates a multi-dimensional narrative quality assessment mechanism based on positive and negative prompts, a Mixture-of-Experts generator, an Aspect Rating Reward Model, Direct Preference Optimization, and an agent-based evaluator. Empirical results demonstrate that the proposed approach outperforms state-of-the-art large language models up to 200 times its size across multiple narrative dimensions and effectively discriminates between high- and low-quality script generations in terms of audience reception.

Technology Category

• Natural Language Processing: Generation
• Humans and AI: Game Design — Procedural Content Generation & Storytelling
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

Creative plot generation presents a fundamental challenge for language models: transforming a concise premise into a coherent narrative that sustains global structure, character development, and emotional resonance. Although recent Large Language Models (LLMs) demonstrate strong fluency across general-purpose tasks, they typically require preference alignment to perform well on specialized domains such as creative plot generation. However, conducting such alignment at the scale of frontier LLMs is computationally prohibitive, significantly limiting accessibility and practical deployment. To address this, we present PlotTwist, a structured framework that enables Small Language Models (SLMs) with $\leq$ 5B active parameters to generate high-quality, premise-conditioned plots competitive with frontier systems up to $200\times$ larger. Our approach decomposes generation into three specialized components: (1) an Aspect Rating Reward Model trained via a novel Positive-Negative prompting strategy to deliver structured narratives across five Narrative Quality Dimensions (NQDs); (2) a Mixture-of-Experts (MoE) plot generator aligned via Direct Preference Optimization on high-confidence preference pairs; and (3) an Agentic Evaluation module that emulates human critical judgment for unbiased post-hoc assessment. Extensive experiments demonstrate that PlotTwist consistently outperforms frontier models across multiple NQDs despite substantially tighter capacity constraints. Further validation confirms strong sensitivity to narrative quality, as the framework reliably distinguishes plots derived from critically acclaimed versus widely panned screenplays. Together, these results establish structured, preference-based alignment as a resource-efficient approach to high-quality creative plot generation.

Problem

• creative plot generation
• small language models
• preference alignment
• narrative coherence
• computational efficiency

Innovation

• Small Language Models
• Preference Alignment
• Mixture-of-Experts
• Narrative Quality Dimensions
• Direct Preference Optimization