Learning Without Losing Identity: Capability Evolution for Embodied Agents
2026-04-09 • Robotics
RoboticsArtificial Intelligence
AI summaryⓘ
The authors propose a new way for robots to get better at tasks without changing who they fundamentally are. Instead of constantly altering the robot itself, they suggest improving separate pieces called Embodied Capability Modules (ECMs), which are like building blocks for skills. These blocks can be updated and combined over time to help the robot do tasks more successfully. Their tests showed this method greatly improved performance while keeping the robot stable and safe. This approach may help robots learn continuously without losing their original identity.
Embodied agentsCapability ModulesModular learningSkill evolutionAgent identityPolicy driftSafety constraintsSkill learning methodsClosed-loop learning
Authors
Xue Qin, Simin Luan, John See, Cong Yang, Zhijun Li
Abstract
Embodied agents are expected to operate persistently in dynamic physical environments, continuously acquiring new capabilities over time. Existing approaches to improving agent performance often rely on modifying the agent itself -- through prompt engineering, policy updates, or structural redesign -- leading to instability and loss of identity in long-lived systems. In this work, we propose a capability-centric evolution paradigm for embodied agents. We argue that a robot should maintain a persistent agent as its cognitive identity, while enabling continuous improvement through the evolution of its capabilities. Specifically, we introduce the concept of Embodied Capability Modules (ECMs), which represent modular, versioned units of embodied functionality that can be learned, refined, and composed over time. We present a unified framework in which capability evolution is decoupled from agent identity. Capabilities evolve through a closed-loop process involving task execution, experience collection, model refinement, and module updating, while all executions are governed by a runtime layer that enforces safety and policy constraints. We demonstrate through simulated embodied tasks that capability evolution improves task success rates from 32.4% to 91.3% over 20 iterations, outperforming both agent-modification baselines and established skill-learning methods (SPiRL, SkiMo), while preserving zero policy drift and zero safety violations. Our results suggest that separating agent identity from capability evolution provides a scalable and safe foundation for long-term embodied intelligence.