# Probing Collision Grounding in Vision-Language Models for Safe Human-Robot Collaboration

arXiv: https://arxiv.org/abs/2605.31196 · HTML: https://mlsanity.com/paper/2605.31196/

## Key facts

- No citing papers in the map yet (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 35 of its 45 references that are in the map.
- First posted on arXiv 29 May 2026.

## 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, 30 of 35)

- [AI2-THOR: An Interactive 3D Environment for Visual AI](https://mlsanity.com/paper/1712.05474/) (2017)
- [BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representation](https://mlsanity.com/paper/2205.13542/) (2022)
- [SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities](https://mlsanity.com/paper/2401.12168/) (2024)
- [Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spaces](https://mlsanity.com/paper/2412.14171/) (2024)
- [Real-world Anomaly Detection in Surveillance Videos](https://mlsanity.com/paper/1801.04264/) (2018)
- [VirtualHome: Simulating Household Activities via Programs](https://mlsanity.com/paper/1806.07011/) (2018)
- [ThreeDWorld: A Platform for Interactive Multi-Modal Physical Simulation](https://mlsanity.com/paper/2007.04954/) (2020)
- [Not only Look, but also Listen: Learning Multimodal Violence Detection under Weak Supervision](https://mlsanity.com/paper/2007.04687/) (2020)
- [Habitat 3.0: A Co-Habitat for Humans, Avatars and Robots](https://mlsanity.com/paper/2310.13724/) (2023)
- [EmbodiedBench: Comprehensive Benchmarking Multi-modal Large Language Models for Vision-Driven Embodied Agents](https://mlsanity.com/paper/2502.09560/) (2025)
- [iGibson 2.0: Object-Centric Simulation for Robot Learning of Everyday Household Tasks](https://mlsanity.com/paper/2108.03272/) (2021)
- [SafeAgentBench: A Benchmark for Safe Task Planning of Embodied LLM Agents](https://mlsanity.com/paper/2412.13178/) (2024)
- [PointVLA: Injecting the 3D World into Vision-Language-Action Models](https://mlsanity.com/paper/2503.07511/) (2025)
- [Harnessing Large Language Models for Training-free Video Anomaly Detection](https://mlsanity.com/paper/2404.01014/) (2024)
- [Holmes-VAU: Towards Long-term Video Anomaly Understanding at Any Granularity](https://mlsanity.com/paper/2412.06171/) (2024)
- [Holmes-VAD: Towards Unbiased and Explainable Video Anomaly Detection via Multi-modal LLM](https://mlsanity.com/paper/2406.12235/) (2024)
- [CORE4D: A 4D Human-Object-Human Interaction Dataset for Collaborative Object REarrangement](https://mlsanity.com/paper/2406.19353/) (2024)
- [IS-Bench: Evaluating Interactive Safety of VLM-Driven Embodied Agents in Daily Household Tasks](https://mlsanity.com/paper/2506.16402/) (2025)
- [AGENTSAFE: Benchmarking the Safety of Embodied Agents on Hazardous Instructions](https://mlsanity.com/paper/2506.14697/) (2025)
- [Generating Robot Constitutions & Benchmarks for Semantic Safety](https://mlsanity.com/paper/2503.08663/) (2025)
- [Agent3D-Zero: An Agent for Zero-shot 3D Understanding](https://mlsanity.com/paper/2403.11835/) (2024)
- [Robobench: A Comprehensive Evaluation Benchmark for Multimodal Large Language Models as Embodied Brain](https://mlsanity.com/paper/2510.17801/) (2025)
- [Customizing Visual-Language Foundation Models for Multi-modal Anomaly Detection and Reasoning](https://mlsanity.com/paper/2403.11083/) (2024)
- [SafeMind: Benchmarking and Mitigating Safety Risks in Embodied LLM Agents](https://mlsanity.com/paper/2509.25885/) (2025)
- [On the Vulnerability of LLM/VLM-Controlled Robotics](https://mlsanity.com/paper/2402.10340/) (2024)
- [Can AI Perceive Physical Danger and Intervene?](https://mlsanity.com/paper/2509.21651/) (2025)
- [Towards Visual Discrimination and Reasoning of Real-World Physical Dynamics: Physics-Grounded Anomaly Detection](https://mlsanity.com/paper/2503.03562/) (2025)
- [Unified Embodied VLM Reasoning with Robotic Action via Autoregressive Discretized Pre-training](https://mlsanity.com/paper/2512.24125/) (2025)
- [R2-Bench: Benchmarking the Robustness of Referring Perception Models under Perturbations](https://mlsanity.com/paper/2403.04924/) (2024)
- [Scalable Benchmarking and Robust Learning for Noise-Free Ego-Motion and 3D Reconstruction from Noisy Video](https://mlsanity.com/paper/2501.14319/) (2025)

## Abstract (arXiv)

Safe human--robot collaboration requires more than visual description: a monitor must determine whether the robot body is safely separated, already colliding with the scene or a person, or about to collide. We call this capability collision grounding: binding visual observations to robot body geometry, camera viewpoint, scene layout, human proximity, and temporal motion in order to infer present and imminent contact. We introduce TouchSafeBench, a physics-grounded benchmark for evaluating collision grounding in vision-language models (VLMs). Built in Habitat~3.0, TouchSafeBench contains 2,940 simulated indoor co-presence episodes across social navigation and social rearrangement, with synchronized multi-view RGB-D observations, top-down trajectory maps, calibrated camera metadata, and simulator-derived contact labels. We study two deployment-facing tasks: classifying the current safety state and warning about imminent collision before contact. Across three frontier or robotics-oriented VLMs and nine visual representations, current models remain far from reliable: the best average Macro-F1 stays below 50%, explicit depth is not automatically transformed into robot-body collision evidence, and robot--scene contact is consistently harder than human-contact risk. TouchSafeBench reveals a central limitation of embodied VLMs: visual fluency does not imply physical accountability. Reliable robot safety monitors will need representations that explicitly bind viewpoint, robot morphology, metric geometry, and future collision. We will release the benchmark upon acceptance.