Breaking Customized LLMs for Coding: Automated Red Teaming for Instruction Backdoor Attacks

LLM customization platforms allow users to build task-specific models for code intelligence tasks by embedding instructions into system prompts, without modifying the underlying model parameters. While these platforms lower the barrier to developing customized LLMs, they also introduce a new atta...

Defenses and Enablers for Skill Injection Attacks on Terminal Based Agents

Large language model LLM agents increasingly rely on reusable skills i.e. documents describing task-specific procedures. However, this introduces a new attack surface for agents to manage. We study two complementary directions for this threat. First, we evaluate guardian-based defenses: an...

DualTAP: A Dual-Task Adversarial Protector for Mobile MLLM Agents

The reliance of mobile GUI agents on Multimodal Large Language Models MLLMs introduces a severe privacy vulnerability: screenshots containing Personally Identifiable Information PII are often sent to untrusted, third-party routers. These routers can exploit their own MLLMs to mine this data,...