Yuqi Jia

AlignSentinel: Alignment-Aware Detection of Prompt Injection Attacks

Explore Data Left Behind in Reinforcement Learning for Reasoning Language Models

Attacking Logic with Logic: Reasoning Injection Attack to Large Reasoning Models

PromptArmor: An Essential Baseline for Prompt Injection Defenses

WAInjectBench: Benchmarking Prompt Injection Detections for Web Agents

Black-Box Red-Teaming of Multi-Agent Systems via Reinforcement Learning

Competitive Advantage Attacks to Decentralized Federated Learning

Tracing Back the Malicious Clients in Poisoning Attacks to Federated Learning

FoundationForensics: Traceback Backdoor Attacks for Vision Foundation Models