LEARNINGCURRENT2026-06-10

Adversarial Attacks on Learned Policies for Surgical Robotic Tasks

Shutong Jin, Ziyang Chen, Preethi Satish, Paavan Gupta, Florian T. Pokorny, Ken Goldberg

Learning-based surgical Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. policies (ACT, Modern Robot LearningDiffusion policyA robot policy that generates actions using diffusion-model techniques., Pi0) can be disrupted by imperceptible visual perturbations, causing 61% Core ConceptsTaskThe job the robot is supposed to complete, such as pick-and-place, navigation, or drawer opening. failure rates in real suturing/debridement experiments. This exposes a critical safety gap: adversarial Modern Robot LearningRobustnessHow well a robot keeps working despite noise, disturbances, or variation. isn't being validated before deploying learned policies in high-stakes medical robotics where patient injury is possible.

THE PROBLEM

This paper focuses on learning. Learning-based surgical Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. policies (ACT, Modern Robot LearningDiffusion policyA robot policy that generates actions using diffusion-model techniques., Pi0) can be disrupted by imperceptible visual perturbations, causing 61% Core ConceptsTaskThe job the robot is supposed to complete, such as pick-and-place, navigation, or drawer opening. failure rates in real suturing/debridement experiments. This exposes a critical safety gap: adversarial Modern Robot LearningRobustnessHow well a robot keeps working despite noise, disturbances, or variation. isn't being validated before deploying learned policies in high-stakes medical robotics where patient injury is possible. Read the paper by tracking the Core ConceptsTaskThe job the robot is supposed to complete, such as pick-and-place, navigation, or drawer opening. definition, the Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. or data assumptions, and the evidence that supports the claimed improvement.

HOW IT WORKS

1

Task framing

The paper frames the work as learning. Start here because it defines what success means and which assumptions the rest of the method inherits.

2

Core method

Learning-based surgical Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. policies (ACT, Modern Robot LearningDiffusion policyA robot policy that generates actions using diffusion-model techniques., Pi0) can be disrupted by imperceptible visual perturbations, causing 61% Core ConceptsTaskThe job the robot is supposed to complete, such as pick-and-place, navigation, or drawer opening. failure rates in real suturing/debridement experiments. This exposes a critical safety gap: adversarial Modern Robot LearningRobustnessHow well a robot keeps working despite noise, disturbances, or variation. isn't being validated before deploying learned policies in high-stakes medical robotics where patient injury is possible. When reading the method section, identify the inputs, the learned or engineered representation, and the Core ConceptsActionA command the robot sends to its motors, controller, or low-level system. or prediction produced by the system.

3

Data and supervision

For robotics work, the data story is part of the method: check whether the system depends on Imitation & Reinforcement LearningTeleoperation (teleop)A human remotely controlling the robot, often to collect demonstrations., Simulation & Sim-to-RealSimulationA virtual environment where robots can be trained or tested., internet video, human labels, or Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. rollouts.

4

Evaluation evidence

The paper should be judged through its Simulation & Sim-to-RealEvaluationMeasuring how well a robot system performs. protocol: what data is used, what Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. or simulator is tested, and which Evaluation & ResearchBaselineA reference method used for comparison. comparisons support the claim. Look for the gap between the headline result and the Simulation & Sim-to-RealDeploymentPutting the trained system on a real robot. setting you would actually care about.

KEY RESULTS

Main contributionConceptual contribution

Learning-based surgical Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. policies (ACT, Modern Robot LearningDiffusion policyA robot policy that generates actions using diffusion-model techniques., Pi0) can be disrupted by imperceptible visual perturbations, causing 61% Core ConceptsTaskThe job the robot is supposed to complete, such as pick-and-place, navigation, or drawer opening. failure rates in real suturing/debridement experiments. This exposes a critical safety gap: adversarial Modern Robot LearningRobustnessHow well a robot keeps working despite noise, disturbances, or variation. isn't being validated before deploying learned policies in high-stakes medical robotics where patient injury is possible.

WHY DEVELOPERS SHOULD CARE

Learning-based surgical Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. policies (ACT, Modern Robot LearningDiffusion policyA robot policy that generates actions using diffusion-model techniques., Pi0) can be disrupted by imperceptible visual perturbations, causing 61% Core ConceptsTaskThe job the robot is supposed to complete, such as pick-and-place, navigation, or drawer opening. failure rates in real suturing/debridement experiments. This exposes a critical safety gap: adversarial Modern Robot LearningRobustnessHow well a robot keeps working despite noise, disturbances, or variation. isn't being validated before deploying learned policies in high-stakes medical robotics where patient injury is possible.

LIMITATIONS

The main limitation to check is whether the claimed behavior holds outside the paper's reported setup. That means testing across different Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. embodiments, scenes, objects, and data distributions.

WHAT COMES NEXT

The practical next step is independent reproduction with clear baselines, ablations, and stress tests. For a developer, the useful follow-up is to map the paper's learning assumptions onto a concrete Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. stack, then test the smallest version of the method that could run end to end.

RELATED PAPERS