# Sim2Val: Leveraging Correlation Across Test Platforms for Variance-Reduced Metric Estimation

arXiv: https://arxiv.org/abs/2506.20553 · HTML: https://mlsanity.com/paper/2506.20553/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Jun 2025.

## Built on by (most cited first, 5 of 5)

- [Reliable and Scalable Robot Policy Evaluation with Imperfect Simulators](https://mlsanity.com/paper/2510.04354/) (2025)
- [Dash2Sim: Closed-Loop Driving Simulation from in-the-wild Dashcam Videos](https://mlsanity.com/paper/2606.07366/) (2026)
- [Coverage Aware Active Evaluation for Failure Discovery with Paired Systems](https://mlsanity.com/paper/2608.13719/) (2026)
- [SCAPE: Scenario-Conditioned Simulation-Augmented Policy Evaluation](https://mlsanity.com/paper/2608.19425/) (2026)
- [Careful Judge: Safe and Efficient Human-AI Collaborative Decision Making](https://mlsanity.com/paper/2610.09043/) (2026)

## Abstract (arXiv)

Learning-based robotic systems demand rigorous validation to assure reliable performance, but extensive real-world testing is often prohibitively expensive, and if conducted may still yield insufficient data for high-confidence guarantees. In this work we introduce Sim2Val, a general estimation framework that leverages paired data across test platforms, e.g., paired simulation and real-world observations, to achieve better estimates of real-world metrics via the method of control variates. By incorporating cheap and abundant auxiliary measurements (for example, simulator outputs) as control variates for costly real-world samples, our method provably reduces the variance of Monte Carlo estimates and thus requires significantly fewer real-world samples to attain a specified confidence bound on the mean performance. We provide theoretical analysis characterizing the variance and sample-efficiency improvement, and demonstrate empirically in autonomous driving and quadruped robotics settings that our approach achieves high-probability bounds with markedly improved sample efficiency. Our technique can lower the real-world testing burden for validating the performance of the stack, thereby enabling more efficient and cost-effective experimental evaluation of robotic systems.