# ReactHuman: A Physics-Grounded Benchmark for Human-Like Reactive Decision-Making in Embodied Multimodal LLMs

arXiv: https://arxiv.org/abs/2609.10895 · HTML: https://mlsanity.com/paper/2609.10895/

## Key facts

- Cited by 1 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 19 of its 26 references that are in the map.
- First posted on arXiv 9 Sep 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, 19 of 19)

- [RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control](https://mlsanity.com/paper/2307.15818/) (2023)
- [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024)
- [π0: A Vision-Language-Action Flow Model for General Robot Control](https://mlsanity.com/paper/2410.24164/) (2024)
- [PaLM-E: An Embodied Multimodal Language Model](https://mlsanity.com/paper/2303.03378/) (2023)
- [Objaverse: A Universe of Annotated 3D Objects](https://mlsanity.com/paper/2212.08051/) (2022)
- [Habitat: A Platform for Embodied AI Research](https://mlsanity.com/paper/1904.01201/) (2019)
- [AI2-THOR: An Interactive 3D Environment for Visual AI](https://mlsanity.com/paper/1712.05474/) (2017)
- [Habitat 2.0: Training Home Assistants to Rearrange their Habitat](https://mlsanity.com/paper/2106.14405/) (2021)
- [ManiSkill2: A Unified Benchmark for Generalizable Manipulation Skills](https://mlsanity.com/paper/2302.04659/) (2023)
- [Holodeck: Language Guided Generation of 3D Embodied AI Environments](https://mlsanity.com/paper/2312.09067/) (2023)
- [ThreeDWorld: A Platform for Interactive Multi-Modal Physical Simulation](https://mlsanity.com/paper/2007.04954/) (2020)
- [RoboGen: Towards Unleashing Infinite Data for Automated Robot Learning via Generative Simulation](https://mlsanity.com/paper/2311.01455/) (2023)
- [ManiSkill: Generalizable Manipulation Skill Benchmark with Large-Scale Demonstrations](https://mlsanity.com/paper/2107.14483/) (2021)
- [Do generative video models understand physical principles?](https://mlsanity.com/paper/2501.09038/) (2025)
- [PhysBench: Benchmarking and Enhancing Vision-Language Models for Physical World Understanding](https://mlsanity.com/paper/2501.16411/) (2025)
- [Physion: Evaluating Physical Prediction from Vision in Humans and Machines](https://mlsanity.com/paper/2106.08261/) (2021)
- [MaskedMimic: Unified Physics-Based Character Control Through Masked Motion Inpainting](https://mlsanity.com/paper/2409.14393/) (2024)
- [HumanoidBench: Simulated Humanoid Benchmark for Whole-Body Locomotion and Manipulation](https://mlsanity.com/paper/2403.10506/) (2024)
- [Physion++: Evaluating Physical Scene Understanding that Requires Online Inference of Different Physical Properties](https://mlsanity.com/paper/2306.15668/) (2023)

## Built on by (most cited first, 1 of 1)

- [RT-Safe: Benchmarking Agent Safety in Real-Time Embodied Environment](https://mlsanity.com/paper/2610.09294/) (2026)

## Abstract (arXiv)

Reacting to sudden physical hazards (catching a slipping plate, dodging a falling knife) is both a meaningful test of embodied intelligence and a hard requirement for deploying multimodal large language models (MLLMs) as the decision coreof household robots. Existing evaluations, however, probe intuitive physics passively through question answering over videos, or target deliberate, long-horizon tasks such as navigation and rearrangement; none measure whether a model can turn physical understanding into immediate, safety-critical action. We introduce ReactHuman, the first physics-grounded benchmark for human-like reactive decision-making, in which the evaluated MLLM acts as the brain of a simulated humanoid facing sudden household hazards; it spans 17 event families and over 1,000 bit-for-bit reproducible scenes with exact, annotation-free ground truth derived from 240 Hz rigid-body simulation, including adversarial objects whose appearance contradicts their physics (a foam anvil, a steel apple). We further design a five-metric suite that scores each reaction along three axes: reasonable, safe, and physically grounded. We physically execute every committed plan so that decisions have observable consequences. With this harness we evaluate seven representative MLLMs. Results show that reactive safety is far from solved: models mishandle roughly one hazard in three, act from fixed dispositions rather than the observed scene, trust appearance over motion, and miss interception points at meter scale even when the chosen action is correct; none of these failures shrink with model scale. ReactHuman thus offers both a fine-grained diagnosis and a scalable training signal toward physically grounded, safety-aware embodied agents. The benchmark can be found here: https://huggingface.co/datasets/Alan123/reacthuman-benchmark-scaled