{
  "id": 19439,
  "url": "https://arxiv.org/abs/2608.13438v1",
  "title": "ContactGuard: Pre-Contact Execution Monitoring with Action-Conditioned Latent World Models",
  "summary": "Contact-rich manipulation failures are often detected only after the robot has committed to contact. This is especially limiting in wrist-camera setups: close gripper--object views help observe contact, but a poor approach may already push, miss, slip, or disturb the object before conventional detectors react. We introduce \\emph{ContactGuard}, a pre-contact execution monitor for chunked visuomotor policies. Given the policy's planned action chunk, ContactGuard predicts its short-horizon conseque",
  "authors": "Gehan Zheng, Matthew Johnson-Roberson, Weiming Zhi",
  "category": "research",
  "topics": "regulation,agents-autonomy",
  "orgs": null,
  "regions": null,
  "published_at": "2026-08-13T16:25:54.000Z",
  "fetched_at": "2026-08-14T05:10:49.168Z",
  "source_slug": "x-arxiv-cs-ai",
  "source_name": "arXiv cs.AI",
  "source_homepage": "https://arxiv.org/list/cs.AI/recent",
  "ethics_ai_record_url": "https://ethics.ai/record/19439",
  "original_url": "https://arxiv.org/abs/2608.13438v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}