Postdoc, Harvard University
3 papers at NeurIPS 2025
We introduce operator-based interpolants—a novel framework that replaces time variables with linear operators in generative models—enabling a single model to perform multiple tasks without task-specific training.
A framework that dynamically adjusts computational resources for robot controllers based on real-time task difficulty, reducing computation time by 2.6-4.4× while maintaining success rates, using the Stochastic Interpolant (SI) framework.
We introduce a framework for training accelerated, few-step generative models that includes consistency models, shortcut models, and mean flow as special cases.
