{
  "id": 12292319,
  "title": "Selecting Long-Horizon Trajectories for Reliable and Efficient Terminal-Agent Training",
  "url": "https://urgent.news/2026/10/05/selecting-long-horizon-trajectories-for-reliable-and-efficient",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-05T05:35:28.000Z",
  "source": {
    "name": "arXiv cs.AI",
    "slug": "arxiv-cs-ai",
    "url": "https://arxiv.org/abs/2610.05831v1"
  },
  "original_language": "en",
  "account": null,
  "summary": "Terminal agents are commonly trained by imitating long teacher trajectories, yet how much of each trajectory to supervise remains unexplored. We study the \\emph{supervision horizon}, the number of trajectory tokens retained for training, and show that it is a key design axis for reliability and cost. Reliability improves with longer horizons but saturates: on Terminal-Bench, a 12K-token horizon…",
  "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."
}