# Beyond Direct Generation: A Decomposed Approach to Well-Crafted Screenwriting with LLMs

arXiv: https://arxiv.org/abs/2510.23163 · HTML: https://mlsanity.com/paper/2510.23163/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 49 references that are in the map.
- First posted on arXiv 27 Oct 2025.

## Builds on (most cited first, 18 of 18)

- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Re3: Generating Longer Stories With Recursive Reprompting and Revision](https://mlsanity.com/paper/2210.06774/) (2022)
- [PlotMachines: Outline-Conditioned Generation with Dynamic Plot State Tracking](https://mlsanity.com/paper/2004.14967/) (2020)
- [Are Large Language Models Capable of Generating Human-Level Narratives?](https://mlsanity.com/paper/2407.13248/) (2024)
- [Agents' Room: Narrative Generation through Multi-step Collaboration](https://mlsanity.com/paper/2410.02603/) (2024)
- [RLCD: Reinforcement Learning from Contrastive Distillation for Language Model Alignment](https://mlsanity.com/paper/2307.12950/) (2023)
- [IBSEN: Director-Actor Agent Collaboration for Controllable and Interactive Drama Script Generation](https://mlsanity.com/paper/2407.01093/) (2024)
- [Finding Flawed Fictions: Evaluating Complex Reasoning in Language Models via Plot Hole Detection](https://mlsanity.com/paper/2504.11900/) (2025)
- [HoLLMwood: Unleashing the Creativity of Large Language Models in Screenwriting via Role Playing](https://mlsanity.com/paper/2406.11683/) (2024)
- [Improving Pacing in Long-Form Story Planning](https://mlsanity.com/paper/2311.04459/) (2023)
- [EIPE-text: Evaluation-Guided Iterative Plan Extraction for Long-Form Narrative Text Generation](https://mlsanity.com/paper/2310.08185/) (2023)
- [Ex3: Automatic Novel Writing by Extracting, Excelsior and Expanding](https://mlsanity.com/paper/2408.08506/) (2024)
- [End-to-end Story Plot Generator](https://mlsanity.com/paper/2310.08796/) (2023)
- [CML-Bench: A Framework for Evaluating and Enhancing LLM-Powered Movie Scripts Generation](https://mlsanity.com/paper/2510.06231/) (2025)
- [Incorporating Commonsense Knowledge into Story Ending Generation via Heterogeneous Graph Networks](https://mlsanity.com/paper/2201.12538/) (2022)
- [CHAE: Fine-Grained Controllable Story Generation with Characters, Actions and Emotions](https://mlsanity.com/paper/2210.05221/) (2022)
- [R2: A LLM Based Novel-to-Screenplay Generation Framework with Causal Plot Graphs](https://mlsanity.com/paper/2503.15655/) (2025)
- [Returning to the Start: Generating Narratives with Related Endpoints](https://mlsanity.com/paper/2404.00829/) (2024)

## Built on by (most cited first, 1 of 1)

- [DramaDirector: Geometry-Guided Short Drama Generation](https://mlsanity.com/paper/2606.24107/) (2026)

## Abstract (arXiv)

The screenplay serves as the foundation for television production, defining narrative structure, character development, and dialogue. While Large Language Models (LLMs) show great potential in creative writing, direct end-to-end generation approaches often fail to produce well-crafted screenplays. We argue this failure stems from forcing a single model to simultaneously master two disparate capabilities: creative narrative construction and rigid format adherence. The resulting outputs may mimic superficial style but lack the deep structural integrity and storytelling substance required for professional use. To enable LLMs to generate high-quality screenplays, we introduce Dual-Stage Refinement (DSR), a decomposed framework that decouples creative narrative generation from format conversion. The first stage transforms a brief outline into rich, novel-style prose. The second stage refines this narrative into a professionally formatted screenplay. This separation enables the model to specialize in one distinct capability at each stage. A key challenge in implementing DSR is the scarcity of paired outline-to-novel training data. We address this through hybrid data synthesis: reverse synthesis deconstructs existing screenplays into structured inputs, while forward synthesis leverages these inputs to generate high-quality narrative texts as training targets. Blind evaluations by professional screenwriters show that DSR achieves a 75% win rate against strong baselines like Gemini-2.5-Pro and reaches 82.7% of human-level performance. Our work demonstrates that decomposed generation architecture with tailored data synthesis effectively specializes LLMs in complex creative domains.