{
  "id": 457,
  "url": "https://arxiv.org/abs/2606.29713v1",
  "title": "SEVA: Self-Evolving Verification Agent with Process Reward for Fact Attribution",
  "summary": "Hallucination is the reliability bottleneck for LLM-based agents, and fact attribution verifiers are the last line of defense -- yet today's verifiers emit only opaque binary labels, leaving agents unable to self-correct and operators unable to audit. We present SEVA, a structured verification agent that emits evidence alignments, step-by-step reasoning chains, calibrated confidence, and a six-category error diagnosis with actionable fixes. Training such an agent with RL is non-trivial: standard",
  "authors": "Aojie Yuan, Yi Nian, Haiyue Zhang, Zijian Su, Yue Zhao",
  "category": "research",
  "topics": "healthcare,military-security,agents-autonomy,transparency",
  "orgs": null,
  "regions": null,
  "published_at": "2026-06-29T02:37:13.000Z",
  "fetched_at": "2026-07-14T14:14:32.648Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/457",
  "original_url": "https://arxiv.org/abs/2606.29713v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}