Principal Researcher, Microsoft Research
1 paper at NeurIPS 2025
We improve and speed up training-free conditional sampling from diffusion models using an approximation to the Newton method