VLACURRENT2026-05-21

Action with Visual Primitives

Weilong Guo, Yuchen Wang, Renping Zhou, Yunfeng Zhang, Rui Fang, Yue Meng, Wenda Xu, Yuan He, Gao Huang

AVP decouples visual understanding from motor Control & PlanningControlThe method used to make the robot move the way you want. by having a language-vision model emit intermediate 'visual primitives' that guide a specialized Core ConceptsActionA command the robot sends to its motors, controller, or low-level system. network, achieving 27.6% higher success rates on Manipulation & TasksPick-and-placePicking up an object from one location and placing it somewhere else. tasks while requiring less Robot LearningTrainingThe process of fitting a model using data or experience. data and generalizing better to new objects. This architecture solves the inefficiency problem where Core ConceptsActionA command the robot sends to its motors, controller, or low-level system. models wastefully relearn Perception & SensingPerceptionThe process of turning raw sensor data into useful understanding of the world. that the Modern Robot LearningVision-Language Model (VLM)A model that understands both images and text. already understands.

THE PROBLEM

This paper focuses on Modern Robot LearningVision-Language-Action model (VLA)A model that takes images and language as input and outputs robot actions.. AVP decouples visual understanding from motor Control & PlanningControlThe method used to make the robot move the way you want. by having a language-vision model emit intermediate 'visual primitives' that guide a specialized Core ConceptsActionA command the robot sends to its motors, controller, or low-level system. network, achieving 27.6% higher success rates on Manipulation & TasksPick-and-placePicking up an object from one location and placing it somewhere else. tasks while requiring less Robot LearningTrainingThe process of fitting a model using data or experience. data and generalizing better to new objects. This architecture solves the inefficiency problem where Core ConceptsActionA command the robot sends to its motors, controller, or low-level system. models wastefully relearn Perception & SensingPerceptionThe process of turning raw sensor data into useful understanding of the world. that the Modern Robot LearningVision-Language Model (VLM)A model that understands both images and text. already understands. 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 Modern Robot LearningVision-Language-Action model (VLA)A model that takes images and language as input and outputs robot actions.. Start here because it defines what success means and which assumptions the rest of the method inherits.

2

Core method

AVP decouples visual understanding from motor Control & PlanningControlThe method used to make the robot move the way you want. by having a language-vision model emit intermediate 'visual primitives' that guide a specialized Core ConceptsActionA command the robot sends to its motors, controller, or low-level system. network, achieving 27.6% higher success rates on Manipulation & TasksPick-and-placePicking up an object from one location and placing it somewhere else. tasks while requiring less Robot LearningTrainingThe process of fitting a model using data or experience. data and generalizing better to new objects. This architecture solves the inefficiency problem where Core ConceptsActionA command the robot sends to its motors, controller, or low-level system. models wastefully relearn Perception & SensingPerceptionThe process of turning raw sensor data into useful understanding of the world. that the Modern Robot LearningVision-Language Model (VLM)A model that understands both images and text. already understands. 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.

FIGURES

KEY RESULTS

Main contributionConceptual contribution

AVP decouples visual understanding from motor Control & PlanningControlThe method used to make the robot move the way you want. by having a language-vision model emit intermediate 'visual primitives' that guide a specialized Core ConceptsActionA command the robot sends to its motors, controller, or low-level system. network, achieving 27.6% higher success rates on Manipulation & TasksPick-and-placePicking up an object from one location and placing it somewhere else. tasks while requiring less Robot LearningTrainingThe process of fitting a model using data or experience. data and generalizing better to new objects. This architecture solves the inefficiency problem where Core ConceptsActionA command the robot sends to its motors, controller, or low-level system. models wastefully relearn Perception & SensingPerceptionThe process of turning raw sensor data into useful understanding of the world. that the Modern Robot LearningVision-Language Model (VLM)A model that understands both images and text. already understands.

WHY DEVELOPERS SHOULD CARE

AVP decouples visual understanding from motor Control & PlanningControlThe method used to make the robot move the way you want. by having a language-vision model emit intermediate 'visual primitives' that guide a specialized Core ConceptsActionA command the robot sends to its motors, controller, or low-level system. network, achieving 27.6% higher success rates on Manipulation & TasksPick-and-placePicking up an object from one location and placing it somewhere else. tasks while requiring less Robot LearningTrainingThe process of fitting a model using data or experience. data and generalizing better to new objects. This architecture solves the inefficiency problem where Core ConceptsActionA command the robot sends to its motors, controller, or low-level system. models wastefully relearn Perception & SensingPerceptionThe process of turning raw sensor data into useful understanding of the world. that the Modern Robot LearningVision-Language Model (VLM)A model that understands both images and text. already understands.

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 Modern Robot LearningVision-Language-Action model (VLA)A model that takes images and language as input and outputs robot actions. 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