# OptiScene: LLM-driven Indoor Scene Layout Generation via Scaled Human-aligned Data Synthesis and Multi-Stage Preference Optimization

arXiv: https://arxiv.org/abs/2506.07570 · HTML: https://mlsanity.com/paper/2506.07570/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- Placed in Self-improving MLLMs and synthetic data, one of 13 directions in Synthetic data and self-improvement for LLMs.
- Builds on 42 of its 51 references that are in the map.
- First posted on arXiv 9 Jun 2025.

## Placed in

- Synthetic data and self-improvement for LLMs / [Self-improving MLLMs and synthetic data](https://mlsanity.com/synth/self-improving-mllms-and-synthetic-data/)

## Builds on (most cited first, 30 of 42)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Objaverse: A Universe of Annotated 3D Objects](https://mlsanity.com/paper/2212.08051/) (2022)
- [Habitat-Matterport 3D Dataset (HM3D): 1000 Large-scale 3D Environments for Embodied AI](https://mlsanity.com/paper/2109.08238/) (2021)
- [Visual-RFT: Visual Reinforcement Fine-Tuning](https://mlsanity.com/paper/2503.01785/) (2025)
- [LM-Nav: Robotic Navigation with Large Pre-Trained Models of Language, Vision, and Action](https://mlsanity.com/paper/2207.04429/) (2022)
- [3D-FRONT: 3D Furnished Rooms with layOuts and semaNTics](https://mlsanity.com/paper/2011.09127/) (2020)
- [ProcTHOR: Large-Scale Embodied AI Using Procedural Generation](https://mlsanity.com/paper/2206.06994/) (2022)
- [LayoutGPT: Compositional Visual Planning and Generation with Large Language Models](https://mlsanity.com/paper/2305.15393/) (2023)
- [NaVid: Video-based VLM Plans the Next Step for Vision-and-Language Navigation](https://mlsanity.com/paper/2402.15852/) (2024)
- [MVImgNet: A Large-scale Dataset of Multi-view Images](https://mlsanity.com/paper/2303.06042/) (2023)
- [ATISS: Autoregressive Transformers for Indoor Scene Synthesis](https://mlsanity.com/paper/2110.03675/) (2021)
- [DiffuScene: Denoising Diffusion Models for Generative Indoor Scene Synthesis](https://mlsanity.com/paper/2303.14207/) (2023)
- [SceneFormer: Indoor Scene Generation with Transformers](https://mlsanity.com/paper/2012.09793/) (2020)
- [HUMANISE: Language-conditioned Human Motion Generation in 3D Scenes](https://mlsanity.com/paper/2210.09729/) (2022)
- [DDFM: Denoising Diffusion Model for Multi-Modality Image Fusion](https://mlsanity.com/paper/2303.06840/) (2023)
- [SceneCraft: An LLM Agent for Synthesizing 3D Scene as Blender Code](https://mlsanity.com/paper/2403.01248/) (2024)
- [PhyScene: Physically Interactable 3D Scene Synthesis for Embodied AI](https://mlsanity.com/paper/2404.09465/) (2024)
- [InstructScene: Instruction-Driven 3D Indoor Scene Synthesis with Semantic Graph Prior](https://mlsanity.com/paper/2402.04717/) (2024)
- [Synthesizing Diverse Human Motions in 3D Indoor Scenes](https://mlsanity.com/paper/2305.12411/) (2023)
- [GraphDreamer: Compositional 3D Scene Synthesis from Scene Graphs](https://mlsanity.com/paper/2312.00093/) (2023)
- [LayoutVLM: Differentiable Optimization of 3D Layout via Vision-Language Models](https://mlsanity.com/paper/2412.02193/) (2024)
- [Automated Creation of Digital Cousins for Robust Policy Learning](https://mlsanity.com/paper/2410.07408/) (2024)
- [I-Design: Personalized LLM Interior Designer](https://mlsanity.com/paper/2404.02838/) (2024)
- [LEGO-Net: Learning Regular Rearrangements of Objects in Rooms](https://mlsanity.com/paper/2301.09629/) (2023)
- [RoboTwin: Dual-Arm Robot Benchmark with Generative Digital Twins (early version)](https://mlsanity.com/paper/2409.02920/) (2024)
- [MIME: Human-Aware 3D Scene Generation](https://mlsanity.com/paper/2212.04360/) (2022)
- [MetaSpatial: Reinforcing 3D Spatial Reasoning in VLMs for the Metaverse](https://mlsanity.com/paper/2503.18470/) (2025)

## Built on by (most cited first, 14 of 14)

- [WorldClaw: Agentic 3D Open-World Generation at Scale](https://mlsanity.com/paper/2608.05248/) (2026)
- [Code-as-Room: Generating 3D Rooms from Top-Down View Images via Agentic Code Synthesis](https://mlsanity.com/paper/2605.18451/) (2026)
- [IL3D: A Large-Scale Indoor Layout Dataset for LLM-Driven 3D Scene Generation](https://mlsanity.com/paper/2510.12095/) (2025)
- [ArtiWorld: LLM-Driven Articulation of 3D Objects in Scenes](https://mlsanity.com/paper/2511.12977/) (2025)
- [Repurposing 3D Generative Model for Autoregressive Layout Generation](https://mlsanity.com/paper/2604.16299/) (2026)
- [STABLE: Simulation-Ready Tabletop Layout Generation via a Semantics-Physics Dual System](https://mlsanity.com/paper/2605.16137/) (2026)
- [SceneCritic: A Symbolic Evaluator for 3D Indoor Scene Synthesis](https://mlsanity.com/paper/2604.13035/) (2026)
- [3D Space as a Scratchpad for Editable Text-to-Image Generation](https://mlsanity.com/paper/2601.14602/) (2026)
- [3D Generation for Embodied AI and Robotic Simulation: A Survey](https://mlsanity.com/paper/2604.26509/) (2026)
- [Function2Scene: 3D Indoor Scene Layout from Functional Specifications](https://mlsanity.com/paper/2605.30819/) (2026)
- [FuncRoom-Agent: Sequential Feed-Forward 3D Functional Indoor Scene Generation](https://mlsanity.com/paper/2608.29519/) (2026)
- [Orchestrating Spatial Semantics via a Zone-Graph Paradigm for Intricate Indoor Scene Generation](https://mlsanity.com/paper/2605.02537/) (2026)
- [Architect-Ant: Editable Automatic Furnishing of Architectural Floor Plans](https://mlsanity.com/paper/2606.10953/) (2026)
- [Learning an Interior Layout Policy in a Domain Specific Language Action Space](https://mlsanity.com/paper/2608.07547/) (2026)

## Abstract (arXiv)

Automatic indoor layout generation has attracted increasing attention due to its potential in interior design, virtual environment construction, and embodied AI. Existing methods fall into two categories: prompt-driven approaches that leverage proprietary LLM services (e.g., GPT APIs) and learning-based methods trained on layout data upon diffusion-based models. Prompt-driven methods often suffer from spatial inconsistency and high computational costs, while learning-based methods are typically constrained by coarse relational graphs and limited datasets, restricting their generalization to diverse room categories. In this paper, we revisit LLM-based indoor layout generation and present 3D-SynthPlace, a large-scale dataset that combines synthetic layouts generated via a 'GPT synthesize, Human inspect' pipeline, upgraded from the 3D-Front dataset. 3D-SynthPlace contains nearly 17,000 scenes, covering four common room types -- bedroom, living room, kitchen, and bathroom -- enriched with diverse objects and high-level spatial annotations. We further introduce OptiScene, a strong open-source LLM optimized for indoor layout generation, fine-tuned based on our 3D-SynthPlace dataset through our two-stage training. For the warum-up stage I, we adopt supervised fine-tuning (SFT), which is taught to first generate high-level spatial descriptions then conditionally predict concrete object placements. For the reinforcing stage II, to better align the generated layouts with human design preferences, we apply multi-turn direct preference optimization (DPO), which significantly improving layout quality and generation success rates. Extensive experiments demonstrate that OptiScene outperforms traditional prompt-driven and learning-based baselines. Moreover, OptiScene shows promising potential in interactive tasks such as scene editing and robot navigation.