We are a research group in the Department of Computer Science at Columbia University, working on the safety, control, and efficacy of neural language systems by deeply understanding the underlying neural networks.

Members

Faculty

PhD Students

Master's & Undergraduate Students

Visitors

Blog

• Improving Parametric Knowledge Access in Reasoning Language Models
• Subliminal Steering: Stronger Encoding of Hidden Signals

Policies

• Lab Policy on AI in Writing & Communication

Joining the lab

Prospective PhD students

Apply to the Columbia Computer Science PhD program and include me in the list of professors you're interested in. I plan to admit roughly two students in the upcoming cycle (for students starting Fall 2026).

Current Columbia PhD students

If you are a current Columbia PhD student interested in collaborating, please email me and I'll do my best to find time to meet.

Undergraduates and master's students

I hope to work with 2–4 undergrad and master's students at any one time. Fill out the interest form to get in touch. It helps, but is not required, to have taken NLP courses.