Zhibo Jin

Graduate Research Student

School of Computer Science

FUNDED RESEARCH

  • Adversarial attacks and robustness testing
  • Model interpretability and attribution methods
  • Trustworthy foundation models (e.g., CLIP and multimodal AI)