Riemann-1.0: An Embodied World Action Model for Physical AI
Haofeng Sun, Jiangbo Pei, Fei Kang, Zexiang Liu, Yaokun Li, Boyi Jiang, Hua Xue, Cindy Zhou, Wei Li, Yichen Wei, Mengyin An, Fanliang Zhao, Biao Jiang, Zile Wang, Yang Liu, Yangguang Li
THE PROBLEM
This paper focuses on world models. Riemann-1.0 unifies Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. Core ConceptsPolicyThe rule or model that maps observations or states to actions. Core ConceptsExecutionActually carrying out planned or predicted actions on the robot. and world Simulation & Sim-to-RealSimulationA virtual environment where robots can be trained or tested. in a single causal autoregressive model trained on 200K+ hours of diverse embodied data, achieving 85% success on long-horizon real-world Manipulation & TasksManipulationUsing a robot arm or hand to move or interact with objects. tasks—this means developers can now leverage pre-trained embodied world models that work across different Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. morphologies without task-specific Modern Robot LearningFine-tuningTaking a pretrained model and adapting it to a specific robot or task.. 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
Task framing
Core method
Data and supervision
Evaluation evidence
KEY RESULTS
Riemann-1.0 unifies Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. Core ConceptsPolicyThe rule or model that maps observations or states to actions. Core ConceptsExecutionActually carrying out planned or predicted actions on the robot. and world Simulation & Sim-to-RealSimulationA virtual environment where robots can be trained or tested. in a single causal autoregressive model trained on 200K+ hours of diverse embodied data, achieving 85% success on long-horizon real-world Manipulation & TasksManipulationUsing a robot arm or hand to move or interact with objects. tasks—this means developers can now leverage pre-trained embodied world models that work across different Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. morphologies without task-specific Modern Robot LearningFine-tuningTaking a pretrained model and adapting it to a specific robot or task..
WHY DEVELOPERS SHOULD CARE
Riemann-1.0 unifies Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. Core ConceptsPolicyThe rule or model that maps observations or states to actions. Core ConceptsExecutionActually carrying out planned or predicted actions on the robot. and world Simulation & Sim-to-RealSimulationA virtual environment where robots can be trained or tested. in a single causal autoregressive model trained on 200K+ hours of diverse embodied data, achieving 85% success on long-horizon real-world Manipulation & TasksManipulationUsing a robot arm or hand to move or interact with objects. tasks—this means developers can now leverage pre-trained embodied world models that work across different Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. morphologies without task-specific Modern Robot LearningFine-tuningTaking a pretrained model and adapting it to a specific robot or task..
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 world models 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.