# SpatialStack: Layered Geometry-Language Fusion for 3D VLM Spatial Reasoning

arXiv: https://arxiv.org/abs/2603.27437 · HTML: https://mlsanity.com/paper/2603.27437/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning.
- Builds on 22 of its 67 references that are in the map.
- First posted on arXiv 28 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, 22 of 22)

- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Models](https://mlsanity.com/paper/2304.10592/) (2023)
- [LLaVA-Video: Video Instruction Tuning With Synthetic Data](https://mlsanity.com/paper/2410.02713/) (2024)
- [ARKitScenes: A Diverse Real-World Dataset For 3D Indoor Scene Understanding Using Mobile RGB-D Data](https://mlsanity.com/paper/2111.08897/) (2021)
- [An Embodied Generalist Agent in 3D World](https://mlsanity.com/paper/2311.12871/) (2023)
- [Mini-Gemini: Mining the Potential of Multi-modality Vision Language Models](https://mlsanity.com/paper/2403.18814/) (2024)
- [LongVILA: Scaling Long-Context Visual Language Models for Long Videos](https://mlsanity.com/paper/2408.10188/) (2024)
- [ConceptFusion: Open-set Multimodal 3D Mapping](https://mlsanity.com/paper/2302.07241/) (2023)
- [What's "up" with vision-language models? Investigating their struggle with spatial reasoning](https://mlsanity.com/paper/2310.19785/) (2023)
- [From Flatland to Space: Teaching Vision-Language Models to Perceive and Reason in 3D](https://mlsanity.com/paper/2503.22976/) (2025)
- [Cambrian-S: Towards Spatial Supersensing in Video](https://mlsanity.com/paper/2511.04670/) (2025)
- [SAT: Dynamic Spatial Aptitude Training for Multimodal Language Models](https://mlsanity.com/paper/2412.07755/) (2024)
- [From CLIP to DINO: Visual Encoders Shout in Multi-modal Large Language Models](https://mlsanity.com/paper/2310.08825/) (2023)
- [Visual Spatial Tuning](https://mlsanity.com/paper/2511.05491/) (2025)
- [STI-Bench: Are MLLMs Ready for Precise Spatial-Temporal World Understanding?](https://mlsanity.com/paper/2503.23765/) (2025)
- [SSR: Enhancing Depth Perception in Vision-Language Models via Rationale-Guided Spatial Reasoning](https://mlsanity.com/paper/2505.12448/) (2025)
- [Spatial-SSRL: Enhancing Spatial Understanding via Self-Supervised Reinforcement Learning](https://mlsanity.com/paper/2510.27606/) (2025)
- [Beyond Semantics: Rediscovering Spatial Awareness in Vision-Language Models](https://mlsanity.com/paper/2503.17349/) (2025)
- [Thinking in Dynamics: How Multimodal Large Language Models Perceive, Track, and Reason Dynamics in Physical 4D World](https://mlsanity.com/paper/2603.12746/) (2026)
- [TULIP: Towards Unified Language-Image Pretraining](https://mlsanity.com/paper/2503.15485/) (2025)
- [DynamicVerse: A Physically-Aware Multimodal Framework for 4D World Modeling](https://mlsanity.com/paper/2512.03000/) (2025)

## Built on by (most cited first, 15 of 15)

- [VGGT-Omega](https://mlsanity.com/paper/2605.15195/) (2026)
- [Annotations as Rollouts: Efficient and Scalable Reinforcement Learning for Video MLLMs](https://mlsanity.com/paper/2608.20492/) (2026)
- [Learning Geometric Representations from Videos for Spatial Intelligent Multimodal Large Language Models](https://mlsanity.com/paper/2606.05833/) (2026)
- [Revisiting Local Context for Long-Horizon Streaming 3D Reconstruction](https://mlsanity.com/paper/2608.27529/) (2026)
- [GeoWeaver: Grounding Visual Tokens with Geometric Evidence before Scene Reasoning](https://mlsanity.com/paper/2605.22558/) (2026)
- [Beyond Single Expert: Harmonizing Diverse Visual Priors in MLLMs for Spatial Understanding](https://mlsanity.com/paper/2607.15054/) (2026)
- [Distill What RGB Can Recover: Privileged 3D Evidence for RGB-Only Vision-Language Models](https://mlsanity.com/paper/2608.00110/) (2026)
- [Multi-View Relational Distillation for Spatial Reasoning with Vision-Language Models](https://mlsanity.com/paper/2608.10864/) (2026)
- [DepthEvidence: Unifying Metric Depth Prediction and Geometric Reasoning in Multimodal Language Models](https://mlsanity.com/paper/2609.31103/) (2026)
- [SpatialSpeak: QA-Native Reconstruction with Local and Global Context for Spatial Chain-of-Thought Reasoning](https://mlsanity.com/paper/2609.33616/) (2026)
- [StageVLN: Spatial and Trajectory Auxiliary Guidance for Efficient Vision-Language Navigation](https://mlsanity.com/paper/2610.05664/) (2026)
- [Seek-and-View Reasoning for Multi-View Spatial Understanding](https://mlsanity.com/paper/2610.11810/) (2026)
- [Geospatial Vector Geometry as a Native Spatial Modality for Vision-Language Models](https://mlsanity.com/paper/2610.16769/) (2026)
- [What Makes a Good Visual Representation for Spatial Understanding in MLLMs?](https://mlsanity.com/paper/2610.17844/) (2026)
- [GeoReader: Learning Latent Geometry for Spatial Reasoning in Vision-Language Models](https://mlsanity.com/paper/2610.18694/) (2026)

## Abstract (arXiv)

Large vision-language models (VLMs) still struggle with reliable 3D spatial reasoning, a core capability for embodied and physical AI systems. This limitation arises from their inability to capture fine-grained 3D geometry and spatial relationships. While recent efforts have introduced multi-view geometry transformers into VLMs, they typically fuse only the deep-layer features from vision and geometry encoders, discarding rich hierarchical signals and creating a fundamental bottleneck for spatial understanding. To overcome this, we propose SpatialStack, a general hierarchical fusion framework that progressively aligns vision, geometry, and language representations across the model hierarchy. Moving beyond conventional late-stage vision-geometry fusion, SpatialStack stacks and synchronizes multi-level geometric features with the language backbone, enabling the model to capture both local geometric precision and global contextual semantics. Building upon this framework, we develop VLM-SpatialStack, a model that achieves state-of-the-art performance on multiple 3D spatial reasoning benchmarks. Extensive experiments and ablations demonstrate that our multi-level fusion strategy consistently enhances 3D understanding and generalizes robustly across diverse spatial reasoning tasks, establishing SpatialStack as an effective and extensible design paradigm for vision-language-geometry integration in next-generation multimodal physical AI systems.