# View-on-Graph: Zero-shot 3D Visual Grounding via Vision-Language Reasoning on Scene Graphs

arXiv: https://arxiv.org/abs/2512.09215 · HTML: https://mlsanity.com/paper/2512.09215/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning.
- Builds on 9 of its 50 references that are in the map.
- First posted on arXiv 10 Dec 2025.

## Placed in

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

## Builds on (most cited first, 9 of 9)

- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Mask3D: Mask Transformer for 3D Semantic Instance Segmentation](https://mlsanity.com/paper/2210.03105/) (2022)
- [SayPlan: Grounding Large Language Models using 3D Scene Graphs for Scalable Robot Task Planning](https://mlsanity.com/paper/2307.06135/) (2023)
- [Think-on-Graph: Deep and Responsible Reasoning of Large Language Model on Knowledge Graph](https://mlsanity.com/paper/2307.07697/) (2023)
- [GPT4Scene: Understand 3D Scenes from Videos with Vision-Language Models](https://mlsanity.com/paper/2501.01428/) (2025)
- [Intent3D: 3D Object Detection in RGB-D Scans Based on Human Intention](https://mlsanity.com/paper/2405.18295/) (2024)
- [RayDF: Neural Ray-surface Distance Fields with Multi-view Consistency](https://mlsanity.com/paper/2310.19629/) (2023)
- [From Cognition to Precognition: A Future-Aware Framework for Social Navigation](https://mlsanity.com/paper/2409.13244/) (2024)

## Built on by (most cited first, 4 of 4)

- [Qwen-3D: A Generalist 3D Vision-Language Model for Spatial Understanding](https://mlsanity.com/paper/2608.02980/) (2026)
- [UniGround: Universal 3D Visual Grounding via Training-Free Scene Parsing](https://mlsanity.com/paper/2603.08131/) (2026)
- [CitySTAR: Structured and Topology-Aware Reasoning for Open-Vocabulary Urban 3D Grounding](https://mlsanity.com/paper/2609.19911/) (2026)
- [Urbex: Agentic Spatial Grounding for City-Scale 3D Scenes](https://mlsanity.com/paper/2610.18356/) (2026)

## Abstract (arXiv)

3D visual grounding (3DVG) identifies objects in 3D scenes from language descriptions. Existing zero-shot approaches leverage 2D vision-language models (VLMs) by converting 3D spatial information (SI) into forms amenable to VLM processing, typically as composite inputs such as specified view renderings or video sequences with overlaid object markers. However, this VLM + SI paradigm yields entangled visual representations that compel the VLM to process entire cluttered cues, making it hard to exploit spatial semantic relationships effectively. In this work, we propose a new VLM x SI paradigm that externalizes the 3D SI into a form enabling the VLM to incrementally retrieve only what it needs during reasoning. We instantiate this paradigm with a novel View-on-Graph (VoG) method, which organizes the scene into a multi-modal, multi-layer scene graph and allows the VLM to operate as an active agent that selectively accesses necessary cues as it traverses the scene. This design offers two intrinsic advantages: (i) by structuring 3D context into a spatially and semantically coherent scene graph rather than confounding the VLM with densely entangled visual inputs, it lowers the VLM's reasoning difficulty; and (ii) by actively exploring and reasoning over the scene graph, it naturally produces transparent, step-by-step traces for interpretable 3DVG. Extensive experiments show that VoG achieves state-of-the-art zero-shot performance, establishing structured scene exploration as a promising strategy for advancing zero-shot 3DVG.