{
  "id": 14955,
  "url": "https://arxiv.org/abs/2607.27937v1",
  "title": "From Scoring to Acting: Outcome-Verified Comparative Self-Distillation for LLM Agents",
  "summary": "Recent work on LLM agents is shifting from external capability elicitation to capability internalization, enabling agents to retain useful skills without retrieval at inference time. On-policy self-distillation (OPSD) offers a promising direction, but many existing methods typically supervise students by scoring actions along student-generated trajectories. Such supervision has two limitations: teacher preferences are not validated by environment outcomes, and action-level scores underuse inform",
  "authors": "Xu Xia, Jinghua Piao, Min Yang, Xiaochong Lan, Jiaju Chen, Yong Li",
  "category": "research",
  "topics": "regulation,children-education,agents-autonomy,environment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-07-30T09:47:58.000Z",
  "fetched_at": "2026-07-31T05:10:57.675Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/14955",
  "original_url": "https://arxiv.org/abs/2607.27937v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}