{
  "id": 7457,
  "url": "https://arxiv.org/abs/2603.09053v1",
  "title": "Sim2Act: Robust Simulation-to-Decision Learning via Adversarial Calibration and Group-Relative Perturbation",
  "summary": "Simulation-to-decision learning enables safe policy training in digital environments without risking real-world deployment, and has become essential in mission-critical domains such as supply chains and industrial systems. However, simulators learned from noisy or biased real-world data often exhibit prediction errors in decision-critical regions, leading to unstable action ranking and unreliable policies. Existing approaches either focus on improving average simulation fidelity or adopt conserv",
  "authors": "Hongyu Cao, Jinghan Zhang, Kunpeng Liu, Dongjie Wang, Feng Xia, Haifeng Chen et al.",
  "category": "research",
  "topics": "bias-fairness,regulation,environment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-03-10T00:51:47.000Z",
  "fetched_at": "2026-07-14T16:33:12.393Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/7457",
  "original_url": "https://arxiv.org/abs/2603.09053v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}