The Alignment Problem Under Partial Observability
Scott Emmons [advisor: Stuart J. Russell]

Towards Principled Training and Serving of Large Language Models
Banghua Zhu [advisor: Michael Jordan and Jiantao Jiao]

Towards Vision-LanguageFoundationModels: Limitations,Improvements, and Generalization
Simon Zhai [advisor: Sergey Levine and Yi Ma]

Vulnerabilities of Language Models
Eric Wallace [advisor: Daniel Klein and Dawn Song]