# How Far Are Vision-Language Models from Constructing the Real World? A Benchmark for Physical Generative Reasoning

arXiv: https://arxiv.org/abs/2603.24866 · HTML: https://mlsanity.com/paper/2603.24866/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning.
- Builds on 24 of its 55 references that are in the map.
- First posted on arXiv 25 Mar 2026.

## Placed in

- LLM reasoning / [Spatial and embodied reasoning](https://mlsanity.com/lr/spatial-and-embodied-reasoning/)
- Multimodal LLMs / vision-language models / [Embodied and 3D vision-language models](https://mlsanity.com/mllm/embodied-and-3d-vision-language-models/)

## Builds on (most cited first, 24 of 24)

- [SpaceR: Reinforcing MLLMs in Video Spatial Reasoning](https://mlsanity.com/paper/2504.01805/) (2025)
- [Reinforcing Spatial Reasoning in Vision-Language Models with Interwoven Thinking and Visual Drawing](https://mlsanity.com/paper/2506.09965/) (2025)
- [WorldModelBench: Judging Video Generation Models As World Models](https://mlsanity.com/paper/2502.20694/) (2025)
- [OmniSpatial: Towards Comprehensive Spatial Reasoning Benchmark for Vision Language Models](https://mlsanity.com/paper/2506.03135/) (2025)
- [Cosmos-Transfer1: Conditional World Generation with Adaptive Multimodal Control](https://mlsanity.com/paper/2503.14492/) (2025)
- [MCP-Universe: Benchmarking Large Language Models with Real-World Model Context Protocol Servers](https://mlsanity.com/paper/2508.14704/) (2025)
- [Why Is Spatial Reasoning Hard for VLMs? An Attention Mechanism Perspective on Focus Areas](https://mlsanity.com/paper/2503.01773/) (2025)
- [HunyuanWorld 1.0: Generating Immersive, Explorable, and Interactive 3D Worlds from Words or Pixels](https://mlsanity.com/paper/2507.21809/) (2025)
- [3D and 4D World Modeling: A Survey](https://mlsanity.com/paper/2509.07996/) (2025)
- [Think with 3D: Geometric Imagination Grounded Spatial Reasoning from Limited Views](https://mlsanity.com/paper/2510.18632/) (2025)
- [MetaSpatial: Reinforcing 3D Spatial Reasoning in VLMs for the Metaverse](https://mlsanity.com/paper/2503.18470/) (2025)
- [LEGO-Puzzles: How Good Are MLLMs at Multi-Step Spatial Reasoning?](https://mlsanity.com/paper/2503.19990/) (2025)
- [Vision-as-Inverse-Graphics Agent via Interleaved Multimodal Reasoning](https://mlsanity.com/paper/2601.11109/) (2026)
- [How to Enable LLM with 3D Capacity? A Survey of Spatial Reasoning in LLM](https://mlsanity.com/paper/2504.05786/) (2025)
- [Open3D-VQA: A Benchmark for Comprehensive Spatial Reasoning with Multimodal Large Language Model in Open Space](https://mlsanity.com/paper/2503.11094/) (2025)
- [EmbodiedGen: Towards a Generative 3D World Engine for Embodied Intelligence](https://mlsanity.com/paper/2506.10600/) (2025)
- [Mind the Gap: Benchmarking Spatial Reasoning in Vision-Language Models](https://mlsanity.com/paper/2503.19707/) (2025)
- [Code2Video: A Code-centric Paradigm for Educational Video Generation](https://mlsanity.com/paper/2510.01174/) (2025)
- [Video2Policy: Scaling up Manipulation Tasks in Simulation through Internet Videos](https://mlsanity.com/paper/2502.09886/) (2025)
- [Matrix-Game: Interactive World Foundation Model](https://mlsanity.com/paper/2506.18701/) (2025)
- [FloorplanQA: A Benchmark for Spatial Reasoning in LLMs using Structured Representations](https://mlsanity.com/paper/2507.07644/) (2025)
- [From Charts to Code: A Hierarchical Benchmark for Multimodal Models](https://mlsanity.com/paper/2510.17932/) (2025)
- [JanusCoder: Towards a Foundational Visual-Programmatic Interface for Code Intelligence](https://mlsanity.com/paper/2510.23538/) (2025)
- [CodeDiffuser: Attention-Enhanced Diffusion Policy via VLM-Generated Code for Instruction Ambiguity](https://mlsanity.com/paper/2506.16652/) (2025)

## Built on by (most cited first, 2 of 2)

- [BIM-Edit: Benchmarking Large Language Models for IFC-Based Building Information Modeling](https://mlsanity.com/paper/2606.20146/) (2026)
- [ScaffoldM3C: A Multimodal Sequential Monte Carlo Framework for Generative Stable Construction Planning](https://mlsanity.com/paper/2610.00487/) (2026)

## Abstract (arXiv)

The physical world is not merely visual; it is governed by rigorous structural and procedural constraints. Yet, the evaluation of vision-language models (VLMs) remains heavily skewed toward perceptual realism, prioritizing the generation of visually plausible 3D layouts, shapes, and appearances. Current benchmarks rarely test whether models grasp the step-by-step processes and physical dependencies required to actually build these artifacts, a capability essential for automating design-to-construction pipelines. To address this, we introduce DreamHouse, a novel benchmark for physical generative reasoning: the capacity to synthesize artifacts that concurrently satisfy geometric, structural, constructability, and code-compliance constraints. We ground this benchmark in residential timber-frame construction, a domain with fully codified engineering standards and objectively verifiable correctness. We curate over 26,000 structures spanning 13 architectural styles, ach verified to construction-document standards (LOD 350) and develop a deterministic 10-test structural validation framework. Unlike static benchmarks that assess only final outputs, DreamHouse supports iterative agentic interaction. Models observe intermediate build states, generate construction actions, and receive structured environmental feedback, enabling a fine-grained evaluation of planning, structural reasoning, and self-correction. Extensive experiments with state-of-the-art VLMs reveal substantial capability gaps that are largely invisible on existing leaderboards. These findings establish physical validity as a critical evaluation axis orthogonal to visual realism, highlighting physical generative reasoning as a distinct and underdeveloped frontier in multimodal intelligence. Available at https://luluyuyuyang.github.io/dreamhouse