{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreife52aqt44szailcqwjlcoecaolvsej3rzcnmwyzcdwd6oxlxhu44",
    "uri": "at://did:plc:lna3xh3vagib3abxsatsvfju/app.bsky.feed.post/3mif4cqenr7s2"
  },
  "path": "/nanotechnology-news3/newsid=69039.php",
  "publishedAt": "2026-03-31T18:39:02.000Z",
  "site": "https://www.nanowerk.com",
  "textContent": "Researchers have created a groundbreaking physics-informed machine-learning model that can run molecular simulations for unprecedented lengths of time, even at temperatures as high as 1000 Kelvin.",
  "title": "Ultra-robust machine-learning models capable of stable molecular simulations at extreme temperatures"
}