Kaiyuan Zhang

Membership Inference Attacks Against Fine-tuned Diffusion Language Models

CodeMirage: Stress-Testing AI-Generated Code Detectors Against Production-Level LLMs

TAI3: Testing Agent Integrity in Interpreting User Intent

$\mu$KE: Matryoshka Unstructured Knowledge Editing of Large Language Models

ProSec: Fortifying Code LLMs with Proactive Security Alignment

ASPIRER: Bypassing System Prompts with Permutation-based Backdoors in LLMs

Profiler: Black-box AI-generated Text Origin Detection via Context-aware Inference Pattern Analysis