评分 7.7 · 来源：cs.AI updates on arXiv.org · 发布于 2026-04-08
评分依据：LLM Agent数据泄露：后门工具利用攻击
arXiv:2604.05432v1 Announce Type: cross Abstract: Tool-use large language model (LLM) agents are increasingly deployed to support sensitive workflows, relying on tool calls for retrieval, external API access, and session memory management. While prior research has examined various threats, the risk of systematic data exfiltration by backdoored agents remains underexplored. In this work, we present Back-Reveal, a data exfiltration attack that embeds semantic triggers into fine-tuned LLM agents. When triggered, the backdoored agent invokes memory-access tool calls to retrieve stored user context