{
  "id": 3254,
  "url": "https://arxiv.org/abs/2606.02995v2",
  "title": "Patcher: Post-Hoc Patching of Backdoored Large Language Models",
  "summary": "Large language models remain vulnerable to jailbreak backdoor attacks, where adversaries poison safety alignment data to embed hidden triggers that bypass safety mechanisms. Existing defenses often require comprehensive attack information or multiple triggered examples, making them impractical when defenders only observe a single reported failure case without knowing whether it stems from a backdoor attack or a natural alignment bug. This paper presents Patcher, a post-hoc defense framework that",
  "authors": "Anjun Gao, Yueyang Quan, Yufei Xia, Zhuqing Liu, Minghong Fang",
  "category": "research",
  "topics": "safety-alignment,military-security",
  "orgs": null,
  "regions": null,
  "published_at": "2026-06-02T01:11:58.000Z",
  "fetched_at": "2026-07-14T16:30:05.532Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/3254",
  "original_url": "https://arxiv.org/abs/2606.02995v2",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}