{
  "id": 6159,
  "url": "https://arxiv.org/abs/2604.07799v2",
  "title": "Learning Without Losing Identity: Capability Evolution for Embodied Agents",
  "summary": "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 pers",
  "authors": "Xue Qin, Simin Luan, John See, Cong Yang, Zhijun Li",
  "category": "research",
  "topics": "regulation,agents-autonomy,environment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-04-09T04:51:07.000Z",
  "fetched_at": "2026-07-14T16:32:15.638Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/6159",
  "original_url": "https://arxiv.org/abs/2604.07799v2",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}