My primary research objective is to develop reliable, efficient, and scalable AI systems. Currently, my work focuses on trustworthy machine learning, efficient machine learning, and theoretical foundations of machine learning, particularly addressing:
- ⚔️ Robustness guarantees against adversarial attacks for large models
- 🛡️ Provably defense mechanisms under threat scenarios for large models
- 📈 Trade-offs between trustworthiness, functionality, and efficiency
During my undergraduate research phase, I was interested in AIGC detection and implicit neural representations for point cloud compression.