# Rethinking Demonstration Unlearning in Imitation Learning for Robotics arXiv: https://arxiv.org/abs/2608.20784 · HTML: https://mlsanity.com/paper/2608.20784/ ## Key facts - Cited by 1 papers in the map (data through 14 Oct 2026). - Placed in Imitation learning methods, one of 13 directions in Offline RL and imitation learning from data. - Builds on 35 of its 40 references that are in the map. - First posted on arXiv 21 Aug 2026. ## Placed in - Offline RL and imitation learning from data / [Imitation learning methods](https://mlsanity.com/offlinerl/imitation-learning-methods/) - LLM safety and security / [Machine unlearning](https://mlsanity.com/safety/machine-unlearning/) ## Builds on (most cited first, 30 of 35) - [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021) - [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024) - [Diffusion Policy: Visuomotor Policy Learning via Action Diffusion](https://mlsanity.com/paper/2303.04137/) (2023) - [π0: A Vision-Language-Action Flow Model for General Robot Control](https://mlsanity.com/paper/2410.24164/) (2024) - [Membership Inference Attacks against Machine Learning Models](https://mlsanity.com/paper/1610.05820/) (2016) - [π0.5: a Vision-Language-Action Model with Open-World Generalization](https://mlsanity.com/paper/2504.16054/) (2025) - [Understanding Black-box Predictions via Influence Functions](https://mlsanity.com/paper/1703.04730/) (2017) - [Machine Unlearning](https://mlsanity.com/paper/1912.03817/) (2019) - [Membership Inference Attacks From First Principles](https://mlsanity.com/paper/2112.03570/) (2021) - [What Matters in Learning from Offline Human Demonstrations for Robot Manipulation](https://mlsanity.com/paper/2108.03298/) (2021) - [Eternal Sunshine of the Spotless Net: Selective Forgetting in Deep Networks](https://mlsanity.com/paper/1911.04933/) (2019) - [Certified Data Removal from Machine Learning Models](https://mlsanity.com/paper/1911.03030/) (2019) - [TOFU: A Task of Fictitious Unlearning for LLMs](https://mlsanity.com/paper/2401.06121/) (2024) - [The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning](https://mlsanity.com/paper/2403.03218/) (2024) - [Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning](https://mlsanity.com/paper/2404.05868/) (2024) - [ML-Leaks: Model and Data Independent Membership Inference Attacks and Defenses on Machine Learning Models](https://mlsanity.com/paper/1806.01246/) (2018) - [Making AI Forget You: Data Deletion in Machine Learning](https://mlsanity.com/paper/1907.05012/) (2019) - [Universal Manipulation Interface: In-The-Wild Robot Teaching Without In-The-Wild Robots](https://mlsanity.com/paper/2402.10329/) (2024) - [SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation](https://mlsanity.com/paper/2310.12508/) (2023) - [Who's Harry Potter? Approximate Unlearning in LLMs](https://mlsanity.com/paper/2310.02238/) (2023) - [Remember What You Want to Forget: Algorithms for Machine Unlearning](https://mlsanity.com/paper/2103.03279/) (2021) - [Towards Unbounded Machine Unlearning](https://mlsanity.com/paper/2302.09880/) (2023) - [A Careful Examination of Large Behavior Models for Multitask Dexterous Manipulation](https://mlsanity.com/paper/2507.05331/) (2025) - [Towards Adversarial Evaluations for Inexact Machine Unlearning](https://mlsanity.com/paper/2201.06640/) (2022) - [Deep Regression Unlearning](https://mlsanity.com/paper/2210.08196/) (2022) - [DataMIL: Selecting Data for Robot Imitation Learning with Datamodels](https://mlsanity.com/paper/2505.09603/) (2025) - [CUPID: Curating Data your Robot Loves with Influence Functions](https://mlsanity.com/paper/2506.19121/) (2025) - [Graph Unlearning with Efficient Partial Retraining](https://mlsanity.com/paper/2403.07353/) (2024) - [Is Your Imitation Learning Policy Better than Mine? Policy Comparison with Near-Optimal Stopping](https://mlsanity.com/paper/2503.10966/) (2025) - [VLA-Forget: Vision-Language-Action Unlearning for Embodied Foundation Models](https://mlsanity.com/paper/2604.03956/) (2026) ## Built on by (most cited first, 1 of 1) - [Published Unlearning Numbers Move Per Checkpoint, and Not Because the Removed Data Survives: An Audit of 263 Released Batch-Normalized Checkpoints](https://mlsanity.com/paper/2609.11490/) (2026) ## Abstract (arXiv) Imitation learning for robotics depends on human demonstrations, some of which people may later ask to remove. Retraining without them is the natural reference, but its cost grows with policy and dataset scale, motivating cheaper operators that edit a trained policy. Metrics inherited from machine unlearning, such as forgetting loss or a single membership attack, do not establish what an edit removed from a policy acting in closed loop. We therefore introduce a retrain-calibrated audit that reads demonstration unlearning along two axes: behavior, whether the edited policy acts like one retrained without the removed demonstrations, and evidence, whether an auditor can still detect it was trained on them. The behavior axis measures action divergence to that retrain at matched states, calibrated by a floor built from independent retrains, so a policy at the floor is as close to a retrain as retrains are to each other. The evidence axis applies a per-demonstration membership attack against a retrain null, reporting both its rank and its absolute member-loss level, since rank alone accepts operators that inflate member losses past the null. A conformal test then combines both axes into one hypothesis of joint retrain consistency, against a fleet of independent retrains large enough to reject at conventional significance. Across five preregistered conditions on three real-robot policy classes and two simulation suites, the axes dissociate in both directions on one checkpoint, as an edit may repair task behavior while leaving evidence unchanged, or reduce evidence while moving behavior away from retraining. On the ACT arm, a redirect edit restores blind-scored robot success to 18 of 20 trials.