Unified Control for Inference-Time Guidance of Denoising Diffusion Models

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Goyal, Maurya, Singh, Anuj, Jamali-Rad, Hadi
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909960015708160
author Goyal, Maurya
Singh, Anuj
Jamali-Rad, Hadi
author_facet Goyal, Maurya
Singh, Anuj
Jamali-Rad, Hadi
contents Aligning diffusion model outputs with downstream objectives is essential for improving task-specific performance. Broadly, inference-time training-free approaches for aligning diffusion models can be categorized into two main strategies: sampling-based methods, which explore multiple candidate outputs and select those with higher reward signals, and gradient-guided methods, which use differentiable reward approximations to directly steer the generation process. In this work, we propose a universal algorithm, UniCoDe, which brings together the strengths of sampling and gradient-based guidance into a unified framework. UniCoDe integrates local gradient signals during sampling, thereby addressing the sampling inefficiency inherent in complex reward-based sampling approaches. By cohesively combining these two paradigms, UniCoDe enables more efficient sampling while offering better trade-offs between reward alignment and divergence from the diffusion unconditional prior. Empirical results demonstrate that UniCoDe remains competitive with state-of-the-art baselines across a range of tasks. The code is available at https://github.com/maurya-goyal10/UniCoDe
format Preprint
id arxiv_https___arxiv_org_abs_2512_12339
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Unified Control for Inference-Time Guidance of Denoising Diffusion Models
Goyal, Maurya
Singh, Anuj
Jamali-Rad, Hadi
Computer Vision and Pattern Recognition
Machine Learning
Aligning diffusion model outputs with downstream objectives is essential for improving task-specific performance. Broadly, inference-time training-free approaches for aligning diffusion models can be categorized into two main strategies: sampling-based methods, which explore multiple candidate outputs and select those with higher reward signals, and gradient-guided methods, which use differentiable reward approximations to directly steer the generation process. In this work, we propose a universal algorithm, UniCoDe, which brings together the strengths of sampling and gradient-based guidance into a unified framework. UniCoDe integrates local gradient signals during sampling, thereby addressing the sampling inefficiency inherent in complex reward-based sampling approaches. By cohesively combining these two paradigms, UniCoDe enables more efficient sampling while offering better trade-offs between reward alignment and divergence from the diffusion unconditional prior. Empirical results demonstrate that UniCoDe remains competitive with state-of-the-art baselines across a range of tasks. The code is available at https://github.com/maurya-goyal10/UniCoDe
title Unified Control for Inference-Time Guidance of Denoising Diffusion Models
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2512.12339