Skip to content
Preprint

RoboWorM: Evaluating Movement Realism, Consistency and Acceleration Affects in Embodied Video World Models

Oct 2026 · 0 citations
Computer Science

TL;DR

This work introduces RoboWorM, a comprehensive, deployment-aware benchmark that leverages known hardware priors such as robot kinematic trees and fixed camera configurations, to systematically examine structural and spatiotemporal fidelity, and conducts the first large-scale audit of the quality-efficiency trade-off.

Abstract

Video world models (VWMs) are rapidly emerging as scalable neural simulators for robotic manipulation. However, most benchmarks only evaluate them as open-world video generators, using Vision-Language Models (VLMs) that lack strict 3D spatial and kinematic reasoning. To bridge this gap, we introduce RoboWorM, a comprehensive, deployment-aware benchmark that leverages known hardware priors such as robot kinematic trees and fixed camera configurations, to systematically examine structural and spatiotemporal fidelity. RoboWorM curates 1,605 episodes across single-arm, bimanual, and humanoid morphologies, spanning both Markovian and long-horizon non-Markovian tasks under single- and multi-view setups. To explicitly measure physical consistency, we propose two deterministic, VLM-free metrics: KIVA for intra-view kinematic stability, and RIGIS for multi-view geometric consistency. By exploiting intrinsic robotic invariants, KIVA improves hallucination detection alignment with human annotations by 25.8% relative to the strongest frontier VLMs, while RIGIS achieves a 130% higher Spearman correlation with human preferences. Furthermore, since real-world deployment requires low inference latency, we utilize RoboWorM to conduct the first large-scale audit of the quality-efficiency trade-off. Evaluating 9 state-of-the-art acceleration techniques across 3 leading world models (over 29,000 generated rollouts), we demonstrate how these acceleration methods can retain superficial visual quality while silently compromising deployment-critical physical constraints. Using our metrics, we achieve 2.3$\times$ maximum inference speed-up with minimal quality loss.

View source

Similar papers

#artificial intelligence Preprint Oct 2026

DepthWorld: 3D World Model for Robot Manipulation

This work introduces a calibration pipeline that combines learned stereo depth with a joint factor graph, pooling all episodes collected from the same physical robot to recover its shared kinematic parameters alongside per-scene extrinsics, and trains DepthWorld, a Stable Video Diffusion-based world model that jointly...

Jai Bardhan, Josef Sivic, Vladimír Petrík · 2 citations
#artificial intelligence Preprint Sep 2026

TriWorldBench: A Tri-View Consistency Perspective on Embodied World Models

TRWORLDBENCH is introduced, a benchmark for evaluating embodied world models through synchronized head, left-wrist, and right-wrist videos and uses 19 metrics to assess tri-view consistency, task alignment, physical and 3D coherence, motion quality, temporal consistency, and visual quality.

Xuan-Yi Liu, Hao-Feng Wang, Rui-Qi Li et al. · 2 citations · ⚡1
Preprint Sep 2026

LIBERO-VPro: Benchmarking Closed-Loop Visual Robustness of Robotic Foundation Models

Robotic foundation models achieve impressive performance on standard manipulation benchmarks, yet these evaluations typically assume clean, timely, and consistent visual observations throughout execution. We introduce LIBERO-VPro, a benchmark for systematically evaluating the closed-loop visual robustness of robotic fo...

Hui-Qiong Li, Zhi-Ting Mei, Anirudha Majumdar et al. · 1 citation
Preprint Oct 2026

PhysicsLENS: Diagnosing Physical Property Blindness in Video Generation Models

This work introduces PhysicsLENS, a dataset and benchmark for evaluating plausibility of physical properties grounded in robotics, and shows that plausible-looking videos often ignore the stated property, and that stating the property lowers plausibility only slightly and not significantly.

Isaiah Milkey, Som Sagar, Aditya Taparia et al. · 0 citations
Sep 2026

IronMind: Scaling Humanoid Dexterous Manipulation via Camera-Space Ego-Centric Pretraining

Egocentric human video offers a scalable data source for dexterous manipulation, yet using it to train humanoid robots presents two challenges: (1) an embodiment gap, as human hands differ structurally from robot end-effectors and low-cost egocentric recordings lack the torso kinematics required by conventional retarge...

Hui-Min Pan, Y. Ren, Kun-Peng Song et al. · 0 citations
#artificial intelligence Preprint Oct 2026

WorldFlow: Joint Video and 3D Motion Prediction via Large-Scale Pretraining

WorldFlow achieves the lowest average displacement error (ADE) and final displacement error (FDE) among the evaluated learning-based methods, and is introduced as a generative world model that jointly predicts future video and query-based 3D trajectories.

Yu-Hao Lu, Jiong-Xu Chen, He-Xiang Wei et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.