# YUBI-Real2Sim: 3DGS Scene Verification for Reliable Policy Evaluation

arXiv: https://arxiv.org/abs/2610.17521 · HTML: https://mlsanity.com/paper/2610.17521/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 35 references that are in the map.
- First posted on arXiv 13 Oct 2026.

## Placed in

- Vision-language-action models

## Builds on (most cited first, 25 of 25)

- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [robosuite: A Modular Simulation Framework and Benchmark for Robot Learning](https://mlsanity.com/paper/2009.12293/) (2020)
- [Evaluating Real-World Robot Manipulation Policies in Simulation](https://mlsanity.com/paper/2405.05941/) (2024)
- [Universal Manipulation Interface: In-The-Wild Robot Teaching Without In-The-Wild Robots](https://mlsanity.com/paper/2402.10329/) (2024)
- [Native and Compact Structured Latents for 3D Generation](https://mlsanity.com/paper/2512.14692/) (2025)
- [Reconciling Reality through Simulation: A Real-to-Sim-to-Real Approach for Robust Manipulation](https://mlsanity.com/paper/2403.03949/) (2024)
- [Approximate Convex Decomposition for 3D Meshes with Collision-Aware Concavity and Tree Search](https://mlsanity.com/paper/2205.02961/) (2022)
- [Re3Sim: Generating High-Fidelity Simulation Data via 3D-Photorealistic Real-to-Sim for Robotic Manipulation](https://mlsanity.com/paper/2502.08645/) (2025)
- [PolaRiS: Scalable Real-to-Sim Evaluations for Generalist Robot Policies](https://mlsanity.com/paper/2512.16881/) (2025)
- [Real-to-Sim Robot Policy Evaluation with Gaussian Splatting Simulation of Soft-Body Interactions](https://mlsanity.com/paper/2511.04665/) (2025)
- [Fisheye-GS: Lightweight and Extensible Gaussian Splatting Module for Fisheye Cameras](https://mlsanity.com/paper/2409.04751/) (2024)
- [GSWorld: Closed-Loop Photo-Realistic Simulation Suite for Robotic Manipulation](https://mlsanity.com/paper/2510.20813/) (2025)
- [Jerk-limited Real-time Trajectory Generation with Arbitrary Target States](https://mlsanity.com/paper/2105.04830/) (2021)
- [SimFoundry: Modular and Automated Scene Generation for Policy Learning and Evaluation](https://mlsanity.com/paper/2606.28276/) (2026)
- [Reliable and Scalable Robot Policy Evaluation with Imperfect Simulators](https://mlsanity.com/paper/2510.04354/) (2025)
- [Self-Calibrating Gaussian Splatting for Large Field of View Reconstruction](https://mlsanity.com/paper/2502.09563/) (2025)
- [SC-OmniGS: Self-Calibrating Omnidirectional Gaussian Splatting](https://mlsanity.com/paper/2502.04734/) (2025)
- [RoboMD: Uncovering Robot Vulnerabilities through Semantic Potential Fields](https://mlsanity.com/paper/2412.02818/) (2024)
- [CrossScore: Towards Multi-View Image Evaluation and Scoring](https://mlsanity.com/paper/2404.14409/) (2024)
- [NVS-SQA: Exploring Self-Supervised Quality Representation Learning for Neurally Synthesized Scenes without References](https://mlsanity.com/paper/2501.06488/) (2025)
- [RoboSnap: One-Shot Real-to-Sim Scene Generation for Generalizable Robot Learning and Evaluation](https://mlsanity.com/paper/2607.06699/) (2026)
- [From Blobs to Spokes: High-Fidelity Surface Reconstruction via Oriented Gaussians](https://mlsanity.com/paper/2604.07337/) (2026)
- [YUBI: Yielding Universal Bidigital Interface for Bimanual Dexterous Manipulation at Scale](https://mlsanity.com/paper/2606.10244/) (2026)
- [UMI-Bench 1.0: An Open and Reproducible Real-World Benchmark for Tabletop Robotic Manipulation with UMI Data](https://mlsanity.com/paper/2606.10382/) (2026)

## Abstract (arXiv)

Evaluating VLA policies on real robots is costly, time-consuming, and difficult to reproduce, making real-to-sim simulation essential for scalable evaluation. While 3D Gaussian Splatting (3DGS) enables rapid real-to-sim scene generation, the reconstruction quality required for reliable policy evaluation remains unquantified. In particular, policies relying on wrist-mounted fisheye cameras (e.g., YUBI handheld gripper data) are heavily exposed to visual artifacts (holes, floaters) in wide-angle views, directly degrading task performance. To address this, we present YUBI-Real2Sim, a unified framework for 3DGS-based scene reconstruction and real-to-sim policy evaluation under dual fisheye wrist views. We develop an integrated pipeline combining 200-degree fisheye camera capture, 3DGS training, 3D foundation models for object modeling, and MuJoCo-based physical simulation, reproducing the observed real-world policy ordering in eight of nine scene-task comparisons. Using this framework, we construct a benchmark across two bimanual robot platforms, three scenes, three tasks, and dual pi0.5-based policies. This systematically measures real-to-sim rank agreement and evaluates policy sensitivity against counterfactual 3DGS artifacts (ceiling holes, floaters, warping) and capture disturbances. We further analyze whether novel-view quality scores can indicate policy-ranking reliability under a scene-held-out evaluation protocol. The results show task-dependent performance and limitations of quality-based reliability prediction, highlighting the need to assess reconstruction quality when using Real2Sim for policy evaluation. These results do not yet establish a generally reliable scene-acceptance rule. Project Page: https://yubi-real2sim.airoa.io.