MANIPULATIONCURRENT2026-06-16

EBench: Elemental Diagnosis of Generalist Mobile Manipulation Policies

Ning Gao, Jinliang Zheng, Xing Gao, Haoxiang Ma, Hanqing Wang, Yukai Wang, Jiantong Chen, Zanxin Chen, Shujie Zhang, Mingda Jia, Xuekun Jiang, Zihou Zhu, Xinyu Li, Shuai Wang, Hao Li, Wenzhe Cai, Yuqiang Yang, Xudong Xu, Zhaoyang Lyu, Yao Mu, Tai Wang, Jiangmiao Pang, Jia Zeng, Weinan Zhang, Chunhua Shen

Instead of comparing policies by one number, EBench reveals that models with similar overall success rates can have radically different strengths—one excels at Manipulation & TasksMobile manipulationA robot both moves around and manipulates objects. but fails at delicate tasks, another retains train performance better. This lets developers identify exactly where their learned policies are weak and what needs fixing. You can now diagnose whether a Core ConceptsPolicyThe rule or model that maps observations or states to actions. fails due to Perception & SensingPerceptionThe process of turning raw sensor data into useful understanding of the world., motor Control & PlanningControlThe method used to make the robot move the way you want., or adaptation rather than just knowing it failed.

THE PROBLEM

This paper focuses on Manipulation & TasksManipulationUsing a robot arm or hand to move or interact with objects.. Instead of comparing policies by one number, EBench reveals that models with similar overall success rates can have radically different strengths—one excels at Manipulation & TasksMobile manipulationA robot both moves around and manipulates objects. but fails at delicate tasks, another retains train performance better. This lets developers identify exactly where their learned policies are weak and what needs fixing. You can now diagnose whether a Core ConceptsPolicyThe rule or model that maps observations or states to actions. fails due to Perception & SensingPerceptionThe process of turning raw sensor data into useful understanding of the world., motor Control & PlanningControlThe method used to make the robot move the way you want., or adaptation rather than just knowing it failed. 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 Manipulation & TasksManipulationUsing a robot arm or hand to move or interact with objects.. Start here because it defines what success means and which assumptions the rest of the method inherits.

2

Core method

Instead of comparing policies by one number, EBench reveals that models with similar overall success rates can have radically different strengths—one excels at Manipulation & TasksMobile manipulationA robot both moves around and manipulates objects. but fails at delicate tasks, another retains train performance better. This lets developers identify exactly where their learned policies are weak and what needs fixing. You can now diagnose whether a Core ConceptsPolicyThe rule or model that maps observations or states to actions. fails due to Perception & SensingPerceptionThe process of turning raw sensor data into useful understanding of the world., motor Control & PlanningControlThe method used to make the robot move the way you want., or adaptation rather than just knowing it failed. 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

Instead of comparing policies by one number, EBench reveals that models with similar overall success rates can have radically different strengths—one excels at Manipulation & TasksMobile manipulationA robot both moves around and manipulates objects. but fails at delicate tasks, another retains train performance better. This lets developers identify exactly where their learned policies are weak and what needs fixing. You can now diagnose whether a Core ConceptsPolicyThe rule or model that maps observations or states to actions. fails due to Perception & SensingPerceptionThe process of turning raw sensor data into useful understanding of the world., motor Control & PlanningControlThe method used to make the robot move the way you want., or adaptation rather than just knowing it failed.

WHY DEVELOPERS SHOULD CARE

Instead of comparing policies by one number, EBench reveals that models with similar overall success rates can have radically different strengths—one excels at Manipulation & TasksMobile manipulationA robot both moves around and manipulates objects. but fails at delicate tasks, another retains train performance better. This lets developers identify exactly where their learned policies are weak and what needs fixing. You can now diagnose whether a Core ConceptsPolicyThe rule or model that maps observations or states to actions. fails due to Perception & SensingPerceptionThe process of turning raw sensor data into useful understanding of the world., motor Control & PlanningControlThe method used to make the robot move the way you want., or adaptation rather than just knowing it failed.

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 Manipulation & TasksManipulationUsing a robot arm or hand to move or interact with objects. 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