Research
Let Them Steal: Trapping Large Language Model Extraction Attacks with Knowledge Honeypot
arXiv:2606.15810v2 Announce Type: replace-cross Abstract: Large language models deployed as commercial APIs are vulnerable to model extraction attacks, while existing defenses either act too late or d
arXiv:2606.15810v2 Announce Type: replace-cross Abstract: Large language models deployed as commercial APIs are vulnerable to model extraction attacks, while existing defenses either act too late or degrade utility for legitimate users. We propose extbf{Knowledge Trap}, a defense that redirects extraction attacks toward low-transferability knowledge through a Honeypot Knowledge Graph (HKG) and breadcrumb-guided exploration. Instead of blocking queries or perturbing outputs, Knowledge Trap consumes the attacker's limited query budget on knowledge with negligible downstream utility while preserving benign-user performance. Experiments in medical and financial domains show that Knowledge Trap reduces surrogate Agreement by 6.2% on average without degrading legitimate-user accuracy, outperforming existing defenses that impose measurable user impact. These results suggest that defending knowledge-space traversal is a practical direction for mitigating LLM extraction attacks.
Related
- CoreGuard: Safeguarding Foundational Capabilities of LLMs Against Model Stealing in Edge Deployment
- AI Model Extraction Attacks: Bypassing Single-Client Assumptions in Defenses
Source: arXiv cs.AI | 2026-08-28