{
  "id": 18435921,
  "title": "MAC-v2: Long-horizon model-based value expansion from pixels",
  "url": "https://urgent.news/2026/11/04/mac-v2-long-horizon-model-based-value-expansion-from-pixels",
  "topic": "ai",
  "section": "AI",
  "published": "2026-11-04T23:51:02.000Z",
  "source": {
    "name": "arXiv cs.AI",
    "slug": "arxiv-cs-ai",
    "url": "https://arxiv.org/abs/2611.07245v1"
  },
  "original_language": "en",
  "account": null,
  "summary": "We study whether \\emph{long-horizon model-based value expansion} can serve as an effective framework in pixel-based offline reinforcement learning (RL). Long-horizon model-based value expansion from pixels is substantially harder than state-based environments, as the agent must learn both the representation and its dynamics, and errors from either can accumulate recursively over imagined…",
  "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."
}