Zizhao Hu
PhD student at USC/ISI.

Salvatori Computer Science Center (SAL)
941 Bloom Walk
Los Angeles, CA 90089
I am a CS Ph.D. student at USC, affiliated with the USC ISI MINDS group, and a former researcher at USC iLab. I have also worked at Georgia Tech’s Agile Systems Lab and Photonics Research Group.
My research focuses on VLMs, and multimodal learning, with current projects on adapting and aligning test-time foundation models.
My long-term goal is to develop efficient AGI to automate labor for EVERYONE and tackle humanity’s toughest engineering challenges, including quantum hardware design, aging and cancer solutions, and controllable fusion energy.
news
Feb 15, 2025 | Last Updated |
---|
selected publications
-
-
- Efficient Multimodal Diffusion Models Using Joint Data Infilling with Partially Shared U-NetarXiv preprint arXiv:2311.16488, 2023
- Class-Incremental Learning Using Generative Experience Replay Based on Time-aware RegularizationarXiv preprint arXiv:2310.03898, 2023
- GalilAI: Out-of-Task Distribution Detection using Causal Active Experimentation for Safe Transfer RL2021