# Critical Interval MSE: Toward Reliable Offline Validation for Robot Manipulation Policies

arXiv: https://arxiv.org/abs/2606.29898 · HTML: https://mlsanity.com/paper/2606.29898/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 32 of its 35 references that are in the map.
- First posted on arXiv 29 Jun 2026.

## Builds on (most cited first, 30 of 32)

- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [π0.5: a Vision-Language-Action Model with Open-World Generalization](https://mlsanity.com/paper/2504.16054/) (2025)
- [Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware](https://mlsanity.com/paper/2304.13705/) (2023)
- [LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning](https://mlsanity.com/paper/2306.03310/) (2023)
- [GR00T N1: An Open Foundation Model for Generalist Humanoid Robots](https://mlsanity.com/paper/2503.14734/) (2025)
- [DROID: A Large-Scale In-The-Wild Robot Manipulation Dataset](https://mlsanity.com/paper/2403.12945/) (2024)
- [What Matters in Learning from Offline Human Demonstrations for Robot Manipulation](https://mlsanity.com/paper/2108.03298/) (2021)
- [RLBench: The Robot Learning Benchmark & Learning Environment](https://mlsanity.com/paper/1909.12271/) (2019)
- [Evaluating Real-World Robot Manipulation Policies in Simulation](https://mlsanity.com/paper/2405.05941/) (2024)
- [Implicit Behavioral Cloning](https://mlsanity.com/paper/2109.00137/) (2021)
- [Doubly Robust Off-policy Value Evaluation for Reinforcement Learning](https://mlsanity.com/paper/1511.03722/) (2015)
- [ManiSkill2: A Unified Benchmark for Generalizable Manipulation Skills](https://mlsanity.com/paper/2302.04659/) (2023)
- [JudgeBench: A Benchmark for Evaluating LLM-based Judges](https://mlsanity.com/paper/2410.12784/) (2024)
- [X-VLA: Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Model](https://mlsanity.com/paper/2510.10274/) (2025)
- [Real-Time Execution of Action Chunking Flow Policies](https://mlsanity.com/paper/2506.07339/) (2025)
- [The Surprising Effectiveness of Representation Learning for Visual Imitation](https://mlsanity.com/paper/2112.01511/) (2021)
- [Data Scaling Laws in Imitation Learning for Robotic Manipulation](https://mlsanity.com/paper/2410.18647/) (2024)
- [THE COLOSSEUM: A Benchmark for Evaluating Generalization for Robotic Manipulation](https://mlsanity.com/paper/2402.08191/) (2024)
- [A Careful Examination of Large Behavior Models for Multitask Dexterous Manipulation](https://mlsanity.com/paper/2507.05331/) (2025)
- [Coarse-to-Fine Imitation Learning: Robot Manipulation from a Single Demonstration](https://mlsanity.com/paper/2105.06411/) (2021)
- [Benchmarks for Deep Off-Policy Evaluation](https://mlsanity.com/paper/2103.16596/) (2021)
- [WorldEval: World Model as Real-World Robot Policies Evaluator](https://mlsanity.com/paper/2505.19017/) (2025)
- [RoboArena: Distributed Real-World Evaluation of Generalist Robot Policies](https://mlsanity.com/paper/2506.18123/) (2025)
- [AutoEval: Autonomous Evaluation of Generalist Robot Manipulation Policies in the Real World](https://mlsanity.com/paper/2503.24278/) (2025)
- [Train Offline, Test Online: A Real Robot Learning Benchmark](https://mlsanity.com/paper/2306.00942/) (2023)
- [Keyframe-Focused Visual Imitation Learning](https://mlsanity.com/paper/2106.06452/) (2021)
- [Hyperparameter Selection for Imitation Learning](https://mlsanity.com/paper/2105.12034/) (2021)
- [A Survey on Imitation Learning for Contact-Rich Tasks in Robotics](https://mlsanity.com/paper/2506.13498/) (2025)
- [Tune to Learn: How Controller Gains Shape Robot Policy Learning](https://mlsanity.com/paper/2604.02523/) (2026)
- [ManipulationNet: An Infrastructure for Benchmarking Real-World Robot Manipulation with Physical Skill Challenges and Embodied Multimodal Reasoning](https://mlsanity.com/paper/2603.04363/) (2026)

## Abstract (arXiv)

Real-world evaluation is the gold standard for robot manipulation policies because it tests them against the physical conditions and deployment challenges they are ultimately designed to handle. However, real-world evaluation is also the bottleneck for iterating on robot imitation learning policies: it is costly, difficult to reproduce, and often too sparse to reliably compare nearby model variants. A straightforward proxy for performance is validation loss on expert demonstrations, but this proxy is often poorly correlated with real-world performance. In this paper, we introduce Critical Interval MSE (CI-MSE), an intuitively simple yet effective offline validation metric. CI-MSE uses a zero-shot agent to localize task-critical segments, restricts error computation to those segments, and applies simple action-alignment procedures that better match rollout-time behavior. Across simulation and real-world experiments, CI-MSE generally yields a stronger correlation between validation error and rollout performance than raw MSE. Across 27 distinct policy checkpoints, CI-MSE achieves a Spearman rank correlation of -0.87, closer to the ideal value of -1 than raw MSE's -0.61, demonstrating a significant improvement. Sensitivity analysis shows that our metric is robust across a wide range of hyperparameter settings. We further study the effectiveness of CI-MSE under evaluation distribution shifts and identify practical boundaries for its use. In summary, this paper provides a simple and reliable offline validation tool for accelerating policy iteration. Project webpage: https://ci-mse.github.io/