{
  "id": 3186,
  "url": "https://arxiv.org/abs/2606.04226v1",
  "title": "PerceptTwin: Semantic Scene Reconstruction for Iterative LLM Planning and Verification",
  "summary": "Simulation environments are useful for both robot policy learning and planning verification and validation. Traditionally, the process of creating a simulation was onerous. Creating a bespoke simulation environment for each individual environment that a robot would operate in was simply infeasible. In this work, we introduce PerceptTwin, a fully automatic pipeline that constructs interactive simulations directly from semantic scene representations produced by a robot's perception stack. PerceptT",
  "authors": "Charlie Gauthier, Sacha Morin, Liam Paull",
  "category": "research",
  "topics": "regulation,agents-autonomy,environment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-06-02T21:25:49.000Z",
  "fetched_at": "2026-07-14T16:30:05.528Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/3186",
  "original_url": "https://arxiv.org/abs/2606.04226v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}