{
  "id": 18315,
  "url": "https://arxiv.org/abs/2608.07892v1",
  "title": "Capability-Routed Guard: Defending Large Reasoning Models Against Reasoning-Centric Jailbreaks",
  "summary": "Large reasoning models (LRMs) expose a new safety failure mode: adversarial prompts can manipulate reasoning context, task decomposition, or capability interpretation so that harmful objectives are processed as legitimate reasoning steps. Existing safeguards, including safety reminders, external classifiers, and self-checking wrappers, are often brittle because they either inspect the adversarial prompt directly or ask the target model to perform additional safety reasoning on the same surface t",
  "authors": "Yiyong Liu, Yixin Wu, Jun Sakuma",
  "category": "research",
  "topics": "safety-alignment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-08-08T03:40:06.000Z",
  "fetched_at": "2026-08-11T05:10:37.351Z",
  "source_slug": "x-arxiv-red-teaming-query",
  "source_name": "arXiv red teaming query",
  "source_homepage": "https://arxiv.org/a/redteam",
  "ethics_ai_record_url": "https://ethics.ai/record/18315",
  "original_url": "https://arxiv.org/abs/2608.07892v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}