# Abstract 3D Perception for Spatial Intelligence in Vision-Language Models

arXiv: https://arxiv.org/abs/2511.10946 · HTML: https://mlsanity.com/paper/2511.10946/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Placed in Embodied and 3D vision-language models, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 17 of its 43 references that are in the map.
- First posted on arXiv 14 Nov 2025.

## Placed in

- 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, 17 of 17)

- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [SAM 2: Segment Anything in Images and Videos](https://mlsanity.com/paper/2408.00714/) (2024)
- [VGGT: Visual Geometry Grounded Transformer](https://mlsanity.com/paper/2503.11651/) (2025)
- [BLINK: Multimodal Large Language Models Can See but Not Perceive](https://mlsanity.com/paper/2404.12390/) (2024)
- [Cosmos-Reason1: From Physical Common Sense To Embodied Reasoning](https://mlsanity.com/paper/2503.15558/) (2025)
- [PhysBench: Benchmarking and Enhancing Vision-Language Models for Physical World Understanding](https://mlsanity.com/paper/2501.16411/) (2025)
- [Magma: A Foundation Model for Multimodal AI Agents](https://mlsanity.com/paper/2502.13130/) (2025)
- [SAT: Dynamic Spatial Aptitude Training for Multimodal Language Models](https://mlsanity.com/paper/2412.07755/) (2024)
- [RelTR: Relation Transformer for Scene Graph Generation](https://mlsanity.com/paper/2201.11460/) (2022)
- [RoboBrain 2.0 Technical Report](https://mlsanity.com/paper/2507.02029/) (2025)
- [Hunyuan3D 1.0: A Unified Framework for Text-to-3D and Image-to-3D Generation](https://mlsanity.com/paper/2411.02293/) (2024)
- [VAGEN: Reinforcing World Model Reasoning for Multi-Turn VLM Agents](https://mlsanity.com/paper/2510.16907/) (2025)
- [MindJourney: Test-Time Scaling with World Models for Spatial Reasoning](https://mlsanity.com/paper/2507.12508/) (2025)
- [Perspective-Aware Reasoning in Vision-Language Models via Mental Imagery Simulation](https://mlsanity.com/paper/2504.17207/) (2025)
- [Visual Embodied Brain: Let Multimodal Large Language Models See, Think, and Control in Spaces](https://mlsanity.com/paper/2506.00123/) (2025)
- [Language-Image Models with 3D Understanding](https://mlsanity.com/paper/2405.03685/) (2024)
- [Robix: A Unified Model for Robot Interaction, Reasoning and Planning](https://mlsanity.com/paper/2509.01106/) (2025)

## Built on by (most cited first, 7 of 7)

- [Thinking in Dynamics: How Multimodal Large Language Models Perceive, Track, and Reason Dynamics in Physical 4D World](https://mlsanity.com/paper/2603.12746/) (2026)
- [Allocentric Perceiver: Disentangling Allocentric Reasoning from Egocentric Visual Priors via Frame Instantiation](https://mlsanity.com/paper/2602.05789/) (2026)
- [Thinking with Novel Views: A Systematic Analysis of Generative-Augmented Spatial Intelligence](https://mlsanity.com/paper/2605.10588/) (2026)
- [GeoWorld-VLM: Geometry from World Models for Vision-Language Models](https://mlsanity.com/paper/2605.16713/) (2026)
- [SpatialSV: Internalizing Interpretable 3D Spatial Awareness in MLLMs via Task-Oriented Visual Supervision](https://mlsanity.com/paper/2606.19915/) (2026)
- [Chain of Spatial Thoughts: Modality-Agnostic Spatial Grounding for Vision Language Models](https://mlsanity.com/paper/2608.10278/) (2026)
- [VLM-in-Sandbox: Visual Workspaces for Agentic Visual Reasoning](https://mlsanity.com/paper/2609.24362/) (2026)

## Abstract (arXiv)

Vision-language models (VLMs) struggle with 3D-related tasks such as spatial cognition and physical understanding, which are crucial for real-world applications like robotics and embodied agents. We attribute this to a modality gap between the 3D tasks and the 2D training of VLM, which led to inefficient retrieval of 3D information from 2D input. To bridge this gap, we introduce SandboxVLM, a simple yet effective framework that leverages abstract bounding boxes to encode geometric structure and physical kinematics for VLM. Specifically, we design a 3D Sandbox reconstruction and perception pipeline comprising four stages: generating multi-view priors with abstract control, proxy elevation, multi-view voting and clustering, and 3D-aware reasoning. Evaluated in zero-shot settings across multiple benchmarks and VLM backbones, our approach consistently improves spatial intelligence, achieving an 8.3% gain on SAT Real compared with baseline methods for instance. These results demonstrate that equipping VLMs with a 3D abstraction substantially enhances their 3D reasoning ability without additional training, suggesting new possibilities for general-purpose embodied intelligence.