Zebang Shen

Postdoc@Department of Computer Science, ETHZ - ETH Zurich·瑞士·OpenReview

研究方向

Differential geometry · Riemannian optimization · Classifier guidance · diffusion model · Optimization · Bilevel Optimization · Langevin dynamics · Poincare inequality

6.7

Flow Density Control: Generative Optimization Beyond Entropy-Regularized Fine-Tuning

NeurIPS 2025Spotlight

6.3

Learning to Steer Markovian Agents under Model Uncertainty

ICLR 2025Poster

三作

6.1

Provable Maximum Entropy Manifold Exploration via Diffusion Models

合作者 (19)

Zebang Shen

When Scores Learn Geometry: Rate Separations under the Manifold Hypothesis

Landing with the Score: Riemannian Optimization through Denoising

A Schrödinger Eigenfunction Method for Long-Horizon Stochastic Optimal Control

Scalable Neural Incentive Design with Parameterized Mean-Field Approximation

Flow Density Control: Generative Optimization Beyond Entropy-Regularized Fine-Tuning

Learning to Steer Markovian Agents under Model Uncertainty

Provable Maximum Entropy Manifold Exploration via Diffusion Models