{
  "id": 6112,
  "url": "https://arxiv.org/abs/2604.08628v1",
  "title": "Retrieval Augmented Classification for Confidential Documents",
  "summary": "Unauthorized disclosure of confidential documents demands robust, low-leakage classification. In real work environments, there is a lot of inflow and outflow of documents. To continuously update knowledge, we propose a methodology for classifying confidential documents using Retrieval Augmented Classification (RAC). To confirm this effectiveness, we compare RAC and supervised fine tuning (FT) on the WikiLeaks US Diplomacy corpus under realistic sequence-length constraints. On balanced data, RAC ",
  "authors": "Yeseul E. Chang, Rahul Kailasa, Simon Shim, Byunghoon Oh, Jaewoo Lee",
  "category": "research",
  "topics": "transparency,environment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-04-09T16:13:03.000Z",
  "fetched_at": "2026-07-14T16:32:15.636Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/6112",
  "original_url": "https://arxiv.org/abs/2604.08628v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}