REALM: A Real-to-Sim Validated Benchmark for Generalization in Robotic Manipulation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sedlacek, Martin, Yefanov, Pavlo, Ponimatkin, Georgy, Bardhan, Jai, Pilc, Simon, Fourmy, Mederic, Kazakos, Evangelos, Snoek, Cees G. M., Sivic, Josef, Petrik, Vladimir
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909973677604864
author Sedlacek, Martin
Yefanov, Pavlo
Ponimatkin, Georgy
Bardhan, Jai
Pilc, Simon
Fourmy, Mederic
Kazakos, Evangelos
Snoek, Cees G. M.
Sivic, Josef
Petrik, Vladimir
author_facet Sedlacek, Martin
Yefanov, Pavlo
Ponimatkin, Georgy
Bardhan, Jai
Pilc, Simon
Fourmy, Mederic
Kazakos, Evangelos
Snoek, Cees G. M.
Sivic, Josef
Petrik, Vladimir
contents Vision-Language-Action (VLA) models empower robots to understand and execute tasks described by natural language instructions. However, a key challenge lies in their ability to generalize beyond the specific environments and conditions they were trained on, which is presently difficult and expensive to evaluate in the real-world. To address this gap, we present REALM, a new simulation environment and benchmark designed to evaluate the generalization capabilities of VLA models, with a specific emphasis on establishing a strong correlation between simulated and real-world performance through high-fidelity visuals and aligned robot control. Our environment offers a suite of 15 perturbation factors, 7 manipulation skills, and more than 3,500 objects. Finally, we establish two task sets that form our benchmark and evaluate the π_{0}, π_{0}-FAST, and GR00T N1.5 VLA models, showing that generalization and robustness remain an open challenge. More broadly, we also show that simulation gives us a valuable proxy for the real-world and allows us to systematically probe for and quantify the weaknesses and failure modes of VLAs. Project page: https://martin-sedlacek.com/realm
format Preprint
id arxiv_https___arxiv_org_abs_2512_19562
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle REALM: A Real-to-Sim Validated Benchmark for Generalization in Robotic Manipulation
Sedlacek, Martin
Yefanov, Pavlo
Ponimatkin, Georgy
Bardhan, Jai
Pilc, Simon
Fourmy, Mederic
Kazakos, Evangelos
Snoek, Cees G. M.
Sivic, Josef
Petrik, Vladimir
Robotics
Artificial Intelligence
Vision-Language-Action (VLA) models empower robots to understand and execute tasks described by natural language instructions. However, a key challenge lies in their ability to generalize beyond the specific environments and conditions they were trained on, which is presently difficult and expensive to evaluate in the real-world. To address this gap, we present REALM, a new simulation environment and benchmark designed to evaluate the generalization capabilities of VLA models, with a specific emphasis on establishing a strong correlation between simulated and real-world performance through high-fidelity visuals and aligned robot control. Our environment offers a suite of 15 perturbation factors, 7 manipulation skills, and more than 3,500 objects. Finally, we establish two task sets that form our benchmark and evaluate the π_{0}, π_{0}-FAST, and GR00T N1.5 VLA models, showing that generalization and robustness remain an open challenge. More broadly, we also show that simulation gives us a valuable proxy for the real-world and allows us to systematically probe for and quantify the weaknesses and failure modes of VLAs. Project page: https://martin-sedlacek.com/realm
title REALM: A Real-to-Sim Validated Benchmark for Generalization in Robotic Manipulation
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2512.19562