Hao Zhu

Adversarial training can be made more effective by considering the hierarchical relationships between classes, leading to vision-language models that are more robust to attacks on both specific classes and their broader categories.

Finally, you can puppeteer photorealistic 3D head avatars with independent control over identity, articulation, and nuanced emotional expression.

Diffusion models, typically used for generation, can now efficiently learn causal structures by smoothing gradients and avoiding expensive matrix inversions.

LLMs, impressive as they are, can't juggle multiple users' conflicting needs without dropping balls on privacy, prioritization, and efficiency.