SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2302.02285 · ICML · 2023

ReDi: Efficient Learning-Free Diffusion Inference via Trajectory Retrieval

Lei Li, William Wang, Kexun Zhang, Xianjun Yang

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Diffusion models show promising generation capability for a variety of data. Despite their high generation quality, the inference for diffusion models is still time-consuming due to the numerous sampling iterations required. To accelerate the inference, we propose REDI, a simple yet learning-free Retrieval-based Diffusion sampling framework. From a precomputed knowledge base, REDI retrieves a trajectory similar to the partially generated trajectory at an early stage of generation, skips a large portion of intermediate steps, and continues sampling from a later step in the retrieved trajectory. We theoretically prove that the generation performance of REDI is guaranteed. Our experiments demonstrate that REDI improves the model inference efficiency by 2× speedup. Furthermore, REDI is able to generalize well in zero-shot cross-domain image genreation such as image stylization. The code and demo for REDI is available at https://github.com/ zkx06111/ReDiffusion.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2302.02285")
get_code_for_paper("2302.02285")
have("2302.02285")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2302.02285.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2302.02285.svg)](https://syntology.ai/paper/2302.02285)

Connect an agent — have() is free.