{
  "id": 5838,
  "url": "https://arxiv.org/abs/2604.13645v1",
  "title": "A Mechanistic Analysis of Sim-and-Real Co-Training in Generative Robot Policies",
  "summary": "Co-training, which combines limited in-domain real-world data with abundant surrogate data such as simulation or cross-embodiment robot data, is widely used for training generative robot policies. Despite its empirical success, the mechanisms that determine when and why co-training is effective remain poorly understood. We investigate the mechanism of sim-and-real co-training through theoretical analysis and empirical study, and identify two intrinsic effects governing performance. The first, \\t",
  "authors": "Yu Lei, Minghuan Liu, Abhiram Maddukuri, Zhenyu Jiang, Yuke Zhu",
  "category": "research",
  "topics": "agents-autonomy,finance-investment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-04-15T09:14:43.000Z",
  "fetched_at": "2026-07-14T16:32:02.061Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/5838",
  "original_url": "https://arxiv.org/abs/2604.13645v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}