{
  "id": 19267412,
  "title": "Deploying Physically and Epistemically Compliant Agents by Distilling Belief-Space Task and Motion Planners",
  "url": "https://urgent.news/2026/11/07/deploying-physically-and-epistemically-compliant-agents-by-distilling",
  "topic": "ai",
  "section": "AI",
  "published": "2026-11-07T03:17:09.000Z",
  "source": {
    "name": "arXiv cs.AI",
    "slug": "arxiv-cs-ai",
    "url": "https://arxiv.org/abs/2611.10430v1"
  },
  "original_language": "en",
  "account": null,
  "summary": "In this work, we train agents to act under uncertainty according to specified, belief-dependent rules. Deep reinforcement learning (RL) can in theory train agents to follow rules if encoded into well-designed rewards, but in practice, RL fails to perfectly optimize these rewards, leading to ineffective or noncompliant agent behavior. Belief-space planners by construction produce behavior in line…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}