arXiv cs.AI
Corrupt Plans, Clean Traces: Evading Chain-of-Thought Monitoring with Plan Injection
부패한 계획, 깨끗한 흔적: 플랜 인젝션을 이용한 사고 연쇄 모니터링 회피
No write-up yet — this entry is still just the auto-fetched hook. Aaron's own summary (200–500 words) will replace this note once he's written it.