# Beyond State-as-Action: Exploiting Command-State Discrepancy for Robot Imitation Learning

arXiv: https://arxiv.org/abs/2609.33145 · HTML: https://mlsanity.com/paper/2609.33145/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Robot manipulation imitation learning, one of 13 directions in Offline RL and imitation learning from data.
- Builds on 24 of its 29 references that are in the map.
- First posted on arXiv 27 Sep 2026.

## Placed in

- Offline RL and imitation learning from data / [Robot manipulation imitation learning](https://mlsanity.com/offlinerl/robot-manipulation-imitation-learning/)

## Builds on (most cited first, 24 of 24)

- [Diffusion Policy: Visuomotor Policy Learning via Action Diffusion](https://mlsanity.com/paper/2303.04137/) (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)
- [Octo: An Open-Source Generalist Robot Policy](https://mlsanity.com/paper/2405.12213/) (2024)
- [DROID: A Large-Scale In-The-Wild Robot Manipulation Dataset](https://mlsanity.com/paper/2403.12945/) (2024)
- [BridgeData V2: A Dataset for Robot Learning at Scale](https://mlsanity.com/paper/2308.12952/) (2023)
- [Perceiver-Actor: A Multi-Task Transformer for Robotic Manipulation](https://mlsanity.com/paper/2209.05451/) (2022)
- [Universal Manipulation Interface: In-The-Wild Robot Teaching Without In-The-Wild Robots](https://mlsanity.com/paper/2402.10329/) (2024)
- [GELLO: A General, Low-Cost, and Intuitive Teleoperation Framework for Robot Manipulators](https://mlsanity.com/paper/2309.13037/) (2023)
- [Human-in-the-Loop Imitation Learning using Remote Teleoperation](https://mlsanity.com/paper/2012.06733/) (2020)
- [Robot Learning on the Job: Human-in-the-Loop Autonomy and Learning During Deployment](https://mlsanity.com/paper/2211.08416/) (2022)
- [Reactive Diffusion Policy: Slow-Fast Visual-Tactile Policy Learning for Contact-Rich Manipulation](https://mlsanity.com/paper/2503.02881/) (2025)
- [State-Only Imitation Learning for Dexterous Manipulation](https://mlsanity.com/paper/2004.04650/) (2020)
- [Adaptive Compliance Policy: Learning Approximate Compliance for Diffusion Guided Control](https://mlsanity.com/paper/2410.09309/) (2024)
- [Robot Data Curation with Mutual Information Estimators](https://mlsanity.com/paper/2502.08623/) (2025)
- [ForceMimic: Force-Centric Imitation Learning with Force-Motion Capture System for Contact-Rich Manipulation](https://mlsanity.com/paper/2410.07554/) (2024)
- [DexForce: Extracting Force-informed Actions from Kinesthetic Demonstrations for Dexterous Manipulation](https://mlsanity.com/paper/2501.10356/) (2025)
- [CUPID: Curating Data your Robot Loves with Influence Functions](https://mlsanity.com/paper/2506.19121/) (2025)
- [ForceVLA2: Unleashing Hybrid Force-Position Control with Force Awareness for Contact-Rich Manipulation](https://mlsanity.com/paper/2603.15169/) (2026)
- [Touch begins where vision ends: Generalizable policies for contact-rich manipulation](https://mlsanity.com/paper/2506.13762/) (2025)
- [In-the-Wild Compliant Manipulation with UMI-FT](https://mlsanity.com/paper/2601.09988/) (2026)
- [FILIC: Dual-Loop Force-Guided Imitation Learning with Impedance Torque Control for Contact-Rich Manipulation Tasks](https://mlsanity.com/paper/2509.17053/) (2025)
- [RealDexUMI: A Wearable Universal Manipulation Interface for Dexterous Robot Learning](https://mlsanity.com/paper/2606.06033/) (2026)
- [Bridging Handheld and Teleoperated Supervision for Contact-Rich Manipulation via State-Gated Experts](https://mlsanity.com/paper/2606.26603/) (2026)

## Abstract (arXiv)

Constructing action targets from measured robot motion is an established approach in imitation learning. Under interaction constraints, however, command-state discrepancy may reflect control demands that motion alone does not capture. We investigate when this information matters and how to exploit it. Across three real-robot tasks, task and phase analyses reveal larger supervision gaps under constrained interaction, while selective command retention provides evidence of locally useful command information. Building on these findings, we propose Command-State Discrepancy Weighting (CSDW), which accounts for robot response times and combines subsequent progress, persistent unmet demand, and demand changes into continuous weights for command supervision. The method requires no task-phase annotations or changes to policy architecture or inference. CSDW improves over uniform command supervision on constrained tasks, while methods perform similarly in the less constrained task. Project page: https://seen-e.github.io/CSDW/.