{
  "id": 15900,
  "url": "https://arxiv.org/abs/2608.01772v1",
  "title": "FRAMES: Guarded and Dual-Objective Skill Evolution for Agents in Policy-Governed Enterprise Workflows",
  "summary": "LLM agents increasingly run policy-bound enterprise workflows such as document auditing, where they must apply rules consistently, ground every value, and stay auditable. Improving these agents is hard: operational feedback is sparse and unlabeled, edits to one rule can regress unrelated cases, and accuracy must improve without inflating inference cost or losing auditability. We present FRAMES, a closed-loop framework that cold-starts deployable skills from existing assets and then evolves them",
  "authors": "Xuhui Wang, Ruoqi Shu, Chen Dan, Tianhua Xu, Mengxi Luo, Yanming Mai et al.",
  "category": "research",
  "topics": "regulation,agents-autonomy,transparency",
  "orgs": null,
  "regions": null,
  "published_at": "2026-08-03T06:46:47.000Z",
  "fetched_at": "2026-08-04T05:10:21.797Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/15900",
  "original_url": "https://arxiv.org/abs/2608.01772v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}