Vincent Siu

发表论文 18 篇 · 总被引 104 次 · h-index 6

代表论文

• AgentVigil: Generic Black-Box Red-teaming for Indirect Prompt Injection against LLM Agents (2025 · 被引 46)
• AGENTVIGIL: Automatic Black-Box Red-teaming for Indirect Prompt Injection against LLM Agents (2025 · Conference on Empirical Methods in Natural Language Processing · 被引 34)
• COSMIC: Generalized Refusal Direction Identification in LLM Activations (2025 · Annual Meeting of the Association for Computational Linguistics · 被引 21)
• A Framework for Formalizing LLM Agent Security (2026 · arXiv.org · 被引 17)
• Agents' Last Exam (2026 · arXiv.org · 被引 14)
• Peer-Preservation in Frontier Models (2026 · arXiv.org · 被引 11)