Inference-Time Alignment for Diffusion Models via Doob's Matching
Chang, Jinyuan, Duan, Chenguang, Jiao, Yuling, Xu, Yi, Yang, Jerry Zhijian
Inference-time alignment for diffusion models aims to adapt a pre-trained diffusion model toward a target distribution without retraining the base score network, thereby preserving the generative capacity of the base model while enforcing desired properties at the inference time. A central mechanism for achieving such alignment is guidance, which modifies the sampling dynamics through an additional drift term. In this work, we introduce Doob's matching, a novel framework for guidance estimation grounded in Doob's $h$-transform. Our approach formulates guidance as the gradient of logarithm of an underlying Doob's $h$-function and employs gradient-penalized regression to simultaneously estimate both the $h$-function and its gradient, resulting in a consistent estimator of the guidance. Theoretically, we establish non-asymptotic convergence rates for the estimated guidance. Moreover, we analyze the resulting controllable diffusion processes and prove non-asymptotic convergence guarantees for the generated distributions in the 2-Wasserstein distance.
Jan-13-2026
- Country:
- Asia > China
- Beijing > Beijing (0.04)
- Hubei Province > Wuhan (0.04)
- Sichuan Province > Chengdu (0.04)
- Europe
- France > Hauts-de-France
- Germany (0.04)
- Slovenia > Drava
- Municipality of Benedikt > Benedikt (0.04)
- United Kingdom > England
- Cambridgeshire > Cambridge (0.04)
- North America > United States
- New York > New York County > New York City (0.04)
- Asia > China
- Genre:
- Research Report (0.40)