Carl Vondrick

Carl Vondrick is an Associate Professor of Computer Science at Columbia University. He leads the Perception and Robotics research group focused on creating intelligent systems that understand and interact with the physical world through vision and robotics. His work emphasizes interpretable models, multimodal learning, and robustness.

Education:

• PhD in Computer Science from MIT (2017), advised by Antonio Torralba
• BS in Computer Science from UC Irvine (2011), advised by Deva Ramanan

Research Interests:

• Generative models for physical intelligence
• Interpretable machine learning
• Robotics and embodied AI
• Large-scale video understanding
• AI for scientific discovery

Article Trends: Recent work focuses on neuro-symbolic methods, generative AI safety, and robot learning. Notable contributions include DIFFusion for subtle visual learning, DiSciPLE for scientific discovery, and differentiable robot rendering frameworks.

Awards:

• 2024 PAMI Young Researcher Award
• 2021 NSF CAREER Award

Advising & Grants: Current PhD students (9) and former advisees (4). Active grants from NSF, DARPA, Toyota Research Institute, and Amazon. Program chairs for ICLR 2025/2026.

Labs/Teams: Leads Columbia's Perception Lab and collaborates with industry partners like Google, Cruise, and TRI.