Juhee Kim

发表论文 6 篇 · 总被引 121 次 · h-index 4

代表论文

• Prompt Flow Integrity to Prevent Privilege Escalation in LLM Agents (2025 · arXiv.org · 被引 86)
• CHANCEL: Efficient Multi-client Isolation Under Adversarial Programs (2021 · Network and Distributed System Security Symposium · 被引 43)
• Tiktag: Breaking ARM's Memory Tagging Extension with Speculative Execution (2024 · IEEE Symposium on Security and Privacy · 被引 22)
• PeTAL: Ensuring Access Control Integrity against Data-only Attacks on Linux (2024 · Conference on Computer and Communications Security · 被引 9)
• Agent Data Injection Attacks are Realistic Threats to AI Agents (2026 · arXiv.org · 被引 3)
• DualView: Preventing Indirect Prompt Injection in Personal AI Agents (2026 · arXiv.org · 被引 1)