{
  "id": 1850894,
  "title": "Policy-Invariant Reward Shaping from LLM Feedback: A Framework for Hybrid RL Agents",
  "url": "https://urgent.news/2026/08/18/policy-invariant-reward-shaping-from-llm-feedback-a-framework-for",
  "topic": "ai",
  "section": "AI",
  "published": "2026-08-18T16:55:46.000Z",
  "source": {
    "name": "arXiv cs.AI",
    "slug": "arxiv-cs-ai",
    "url": "https://arxiv.org/abs/2608.18008v1"
  },
  "original_language": "en",
  "account": null,
  "summary": "Combining large language models with reinforcement learning is increasingly explored, yet the theoretical status of LLM-derived reward signals is often left implicit. We formalize the hybrid LLM-planner and RL-controller architecture as a Goal-Augmented Markov Decision Process and show that when the LLM per-state progress score is used as a bounded potential function, the resulting shaping term…",
  "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."
}