# 3D-Generalist: Self-Improving Vision-Language-Action Models for Crafting 3D Worlds

arXiv: https://arxiv.org/abs/2507.06484 · HTML: https://mlsanity.com/paper/2507.06484/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning.
- Builds on 48 of its 97 references that are in the map.
- First posted on arXiv 9 Jul 2025.

## Placed in

- LLM reasoning / [Spatial and embodied reasoning](https://mlsanity.com/lr/spatial-and-embodied-reasoning/)
- 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 48)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Objaverse: A Universe of Annotated 3D Objects](https://mlsanity.com/paper/2212.08051/) (2022)
- [Large Language Models Cannot Self-Correct Reasoning Yet](https://mlsanity.com/paper/2310.01798/) (2023)
- [Grounded SAM: Assembling Open-World Models for Diverse Visual Tasks](https://mlsanity.com/paper/2401.14159/) (2024)
- [LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content Creation](https://mlsanity.com/paper/2402.05054/) (2024)
- [SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities](https://mlsanity.com/paper/2401.12168/) (2024)
- [Object Hallucination in Image Captioning](https://mlsanity.com/paper/1809.02156/) (2018)
- [Hypersim: A Photorealistic Synthetic Dataset for Holistic Indoor Scene Understanding](https://mlsanity.com/paper/2011.02523/) (2020)
- [CLAY: A Controllable Large-scale Generative Model for Creating High-quality 3D Assets](https://mlsanity.com/paper/2406.13897/) (2024)
- [InstantMesh: Efficient 3D Mesh Generation from a Single Image with Sparse-view Large Reconstruction Models](https://mlsanity.com/paper/2404.07191/) (2024)
- [Training Language Models to Self-Correct via Reinforcement Learning](https://mlsanity.com/paper/2409.12917/) (2024)
- [RLHF-V: Towards Trustworthy MLLMs via Behavior Alignment from Fine-grained Correctional Human Feedback](https://mlsanity.com/paper/2312.00849/) (2023)
- [Florence-2: Advancing a Unified Representation for a Variety of Vision Tasks](https://mlsanity.com/paper/2311.06242/) (2023)
- [RoboCasa: Large-Scale Simulation of Everyday Tasks for Generalist Robots](https://mlsanity.com/paper/2406.02523/) (2024)
- [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)
- [AMBER: An LLM-free Multi-dimensional Benchmark for MLLMs Hallucination Evaluation](https://mlsanity.com/paper/2311.07397/) (2023)
- [Holodeck: Language Guided Generation of 3D Embodied AI Environments](https://mlsanity.com/paper/2312.09067/) (2023)
- [Text2Room: Extracting Textured 3D Meshes from 2D Text-to-Image Models](https://mlsanity.com/paper/2303.11989/) (2023)
- [RoboGen: Towards Unleashing Infinite Data for Automated Robot Learning via Generative Simulation](https://mlsanity.com/paper/2311.01455/) (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)
- [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)
- [Is A Picture Worth A Thousand Words? Delving Into Spatial Reasoning for Vision Language Models](https://mlsanity.com/paper/2406.14852/) (2024)
- [URDFormer: A Pipeline for Constructing Articulated Simulation Environments from Real-World Images](https://mlsanity.com/paper/2405.11656/) (2024)
- [GALA3D: Towards Text-to-3D Complex Scene Generation via Layout-guided Generative Gaussian Splatting](https://mlsanity.com/paper/2402.07207/) (2024)

## Built on by (most cited first, 3 of 3)

- [Vision-as-Inverse-Graphics Agent via Interleaved Multimodal Reasoning](https://mlsanity.com/paper/2601.11109/) (2026)
- [SceneSmith: Agentic Generation of Simulation-Ready Indoor Scenes](https://mlsanity.com/paper/2602.09153/) (2026)
- [Transductive Visual Programming: Evolving Tool Libraries from Experience for Spatial Reasoning](https://mlsanity.com/paper/2512.20934/) (2025)

## Abstract (arXiv)

Despite large-scale pretraining endowing models with language and vision reasoning capabilities, improving their spatial reasoning capability remains challenging due to the lack of data grounded in the 3D world. While it is possible for humans to manually create immersive and interactive worlds through 3D graphics, as seen in applications such as VR, gaming, and robotics, this process remains highly labor-intensive. In this paper, we propose a scalable method for generating high-quality 3D environments that can serve as training data for foundation models. We recast 3D environment building as a sequential decision-making problem, employing Vision-Language-Models (VLMs) as policies that output actions to jointly craft a 3D environment's layout, materials, lighting, and assets. Our proposed framework, 3D-Generalist, trains VLMs to generate more prompt-aligned 3D environments via self-improvement fine-tuning. We demonstrate the effectiveness of 3D-Generalist and the proposed training strategy in generating simulation-ready 3D environments. Furthermore, we demonstrate its quality and scalability in synthetic data generation by pretraining a vision foundation model on the generated data. After fine-tuning the pre-trained model on downstream tasks, we show that it surpasses models pre-trained on meticulously human-crafted synthetic data and approaches results achieved with real data orders of magnitude larger.