# TheaterGen: Character Management with LLM for Consistent Multi-turn Image Generation

arXiv: https://arxiv.org/abs/2404.18919 · HTML: https://mlsanity.com/paper/2404.18919/

## Key facts

- Cited by 19 papers in the map (data through 14 Oct 2026).
- Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models.
- Builds on 41 of its 50 references that are in the map.
- First posted on arXiv 29 Apr 2024.

## Placed in

- Diffusion models / [Image generation, text-to-image and editing](https://mlsanity.com/diffusion/image-generation-text-to-image-and-editing/)

## Builds on (most cited first, 30 of 41)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [A Style-Based Generator Architecture for Generative Adversarial Networks](https://mlsanity.com/paper/1812.04948/) (2018)
- [Generative Adversarial Networks](https://mlsanity.com/paper/2203.00667/) (2022)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation](https://mlsanity.com/paper/2201.12086/) (2022)
- [Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation](https://mlsanity.com/paper/1406.1078/) (2014)
- [Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision](https://mlsanity.com/paper/2102.05918/) (2021)
- [Analyzing and Improving the Image Quality of StyleGAN](https://mlsanity.com/paper/1912.04958/) (2019)
- [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://mlsanity.com/paper/2303.05499/) (2023)
- [Tree of Thoughts: Deliberate Problem Solving with Large Language Models](https://mlsanity.com/paper/2305.10601/) (2023)
- [Align before Fuse: Vision and Language Representation Learning with Momentum Distillation](https://mlsanity.com/paper/2107.07651/) (2021)
- [IP-Adapter: Text Compatible Image Prompt Adapter for Text-to-Image Diffusion Models](https://mlsanity.com/paper/2308.06721/) (2023)
- [NExT-GPT: Any-to-Any Multimodal LLM](https://mlsanity.com/paper/2309.05519/) (2023)
- [Scaling up GANs for Text-to-Image Synthesis](https://mlsanity.com/paper/2303.05511/) (2023)
- [Make-A-Scene: Scene-Based Text-to-Image Generation with Human Priors](https://mlsanity.com/paper/2203.13131/) (2022)
- [DreamLLM: Synergistic Multimodal Comprehension and Creation](https://mlsanity.com/paper/2309.11499/) (2023)
- [Generating Images with Multimodal Language Models](https://mlsanity.com/paper/2305.17216/) (2023)
- [LayoutDiffusion: Controllable Diffusion Model for Layout-to-image Generation](https://mlsanity.com/paper/2303.17189/) (2023)
- [LLM-grounded Diffusion: Enhancing Prompt Understanding of Text-to-Image Diffusion Models with Large Language Models](https://mlsanity.com/paper/2305.13655/) (2023)
- [Making LLaMA SEE and Draw with SEED Tokenizer](https://mlsanity.com/paper/2310.01218/) (2023)
- [StoryGAN: A Sequential Conditional GAN for Story Visualization](https://mlsanity.com/paper/1812.02784/) (2018)
- [Training-Free Consistent Text-to-Image Generation](https://mlsanity.com/paper/2402.03286/) (2024)
- [Ranni: Taming Text-to-Image Diffusion for Accurate Instruction Following](https://mlsanity.com/paper/2311.17002/) (2023)
- [TaleCrafter: Interactive Story Visualization with Multiple Characters](https://mlsanity.com/paper/2305.18247/) (2023)
- [Tell, Draw, and Repeat: Generating and Modifying Images Based on Continual Linguistic Instruction](https://mlsanity.com/paper/1811.09845/) (2018)

## Built on by (most cited first, 19 of 19)

- [DreamStory: Open-Domain Story Visualization by LLM-Guided Multi-Subject Consistent Diffusion](https://mlsanity.com/paper/2407.12899/) (2024)
- [AutoStudio: Crafting Consistent Subjects in Multi-turn Interactive Image Generation](https://mlsanity.com/paper/2406.01388/) (2024)
- [AniMaker: Multi-Agent Animated Storytelling with MCTS-Driven Clip Generation](https://mlsanity.com/paper/2506.10540/) (2025)
- [Unbounded: A Generative Infinite Game of Character Life Simulation](https://mlsanity.com/paper/2410.18975/) (2024)
- [Story-Iter: A Training-free Iterative Paradigm for Long Story Visualization](https://mlsanity.com/paper/2410.06244/) (2024)
- [A Character-Centric Creative Story Generation via Imagination](https://mlsanity.com/paper/2409.16667/) (2024)
- [Object Isolated Attention for Consistent Story Visualization](https://mlsanity.com/paper/2503.23353/) (2025)
- [AnimeGamer: Infinite Anime Life Simulation with Next Game State Prediction](https://mlsanity.com/paper/2504.01014/) (2025)
- [ViStoryBench: Comprehensive Benchmark Suite for Story Visualization](https://mlsanity.com/paper/2505.24862/) (2025)
- [VisAgent: Narrative-Preserving Story Visualization Framework](https://mlsanity.com/paper/2503.02399/) (2025)
- [CountLoop: Training-Free High-Instance Image Generation via Iterative Agent Guidance](https://mlsanity.com/paper/2508.16644/) (2025)
- [Improving Multi-Subject Consistency in Open-Domain Image Generation with Isolation and Reposition Attention](https://mlsanity.com/paper/2411.19261/) (2024)
- [CANVAS: Continuity-Aware Narratives via Visual Agentic Storyboarding](https://mlsanity.com/paper/2604.13452/) (2026)
- [Bridging Your Imagination with Audio-Video Generation via a Unified Director](https://mlsanity.com/paper/2512.23222/) (2025)
- [Plot'n Polish: Zero-shot Story Visualization and Disentangled Editing with Text-to-Image Diffusion Models](https://mlsanity.com/paper/2509.04446/) (2025)
- [DreamingComics: A Story Visualization Pipeline via Subject and Layout Customized Generation using Video Models](https://mlsanity.com/paper/2512.01686/) (2025)
- [2K-Characters-10K-Stories: A Quality-Gated Stylized Narrative Dataset with Disentangled Control and Sequence Consistency](https://mlsanity.com/paper/2512.05557/) (2025)
- [MoGen: A Unified Collaborative Framework for Controllable Multi-Object Image Generation](https://mlsanity.com/paper/2601.05546/) (2026)
- [ELECT: Settling Identity Before Rendering for Scalable Story Visualization](https://mlsanity.com/paper/2610.15555/) (2026)

## Abstract (arXiv)

Recent advances in diffusion models can generate high-quality and stunning images from text. However, multi-turn image generation, which is of high demand in real-world scenarios, still faces challenges in maintaining semantic consistency between images and texts, as well as contextual consistency of the same subject across multiple interactive turns. To address this issue, we introduce TheaterGen, a training-free framework that integrates large language models (LLMs) and text-to-image (T2I) models to provide the capability of multi-turn image generation. Within this framework, LLMs, acting as a "Screenwriter", engage in multi-turn interaction, generating and managing a standardized prompt book that encompasses prompts and layout designs for each character in the target image. Based on these, Theatergen generate a list of character images and extract guidance information, akin to the "Rehearsal". Subsequently, through incorporating the prompt book and guidance information into the reverse denoising process of T2I diffusion models, Theatergen generate the final image, as conducting the "Final Performance". With the effective management of prompt books and character images, TheaterGen significantly improves semantic and contextual consistency in synthesized images. Furthermore, we introduce a dedicated benchmark, CMIGBench (Consistent Multi-turn Image Generation Benchmark) with 8000 multi-turn instructions. Different from previous multi-turn benchmarks, CMIGBench does not define characters in advance. Both the tasks of story generation and multi-turn editing are included on CMIGBench for comprehensive evaluation. Extensive experimental results show that TheaterGen outperforms state-of-the-art methods significantly. It raises the performance bar of the cutting-edge Mini DALLE 3 model by 21% in average character-character similarity and 19% in average text-image similarity.