Yiheng Xiong

Mobile GUI agents are surprisingly susceptible to prompt injection via realistic, attacker-controlled text embedded within ordinary user-generated content, even without modifying the agent, application, or OS.

LLMs can now automatically generate property-based tests for Android apps, uncovering 25 previously unknown bugs that eluded existing functional testing techniques.

LLMs can help automated mobile app testing break free from UI "tarpits," leading to 50% higher code coverage and the discovery of dozens of previously unknown bugs in real-world apps.