Inference-Time Scaling for Diffusion Models beyond Scaling Denoising Steps

TL;DR

This paper investigates how increasing inference-time computation, specifically the number of denoising steps, can improve the quality of diffusion model outputs beyond traditional limits, by exploring search strategies for better noise initialization.

Contribution

It introduces a framework for inference-time scaling in diffusion models, including search algorithms for better noise, demonstrating significant quality improvements with increased computation.

Findings

Inference-time compute boosts sample quality.

Combining search strategies and denoising steps enhances performance.

Framework adapts to different application scenarios.

Abstract

Generative models have made significant impacts across various domains, largely due to their ability to scale during training by increasing data, computational resources, and model size, a phenomenon characterized by the scaling laws. Recent research has begun to explore inference-time scaling behavior in Large Language Models (LLMs), revealing how performance can further improve with additional computation during inference. Unlike LLMs, diffusion models inherently possess the flexibility to adjust inference-time computation via the number of denoising steps, although the performance gains typically flatten after a few dozen. In this work, we explore the inference-time scaling behavior of diffusion models beyond increasing denoising steps and investigate how the generation performance can further improve with increased computation. Specifically, we consider a search problem aimed at identifying better noises for the diffusion sampling process. We structure the design space along two axes: the verifiers used to provide feedback, and the algorithms used to find better noise candidates. Through extensive experiments on class-conditioned and text-conditioned image generation benchmarks, our findings reveal that increasing inference-time compute leads to substantial improvements in the quality of samples generated by diffusion models, and with the complicated nature of images, combinations of the components in the framework can be specifically chosen to conform with different application scenario.