SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2310.17347 · ICLR · 2024

CADS: Unleashing the Diversity of Diffusion Models through Condition-Annealed Sampling

Otmar Hilliges, Seyedmorteza Sadat, Jakob Buhmann, Derek Bradley, Romann Weber

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. We have no repository linked to it either. That is a statement about our graph, not about the paper.

Abstract

While conditional diffusion models are known to have good coverage of the data distribution, they still face limitations in output diversity, particularly when sampled with a high classifier-free guidance scale for optimal image quality or when trained on small datasets. We attribute this problem to the role of the conditioning signal in inference and offer an improved sampling strategy for diffusion models that can increase generation diversity, especially at high guidance scales, with minimal loss of sample quality. Our sampling strategy anneals the conditioning signal by adding scheduled, monotonically decreasing Gaussian noise to the conditioning vector during inference to balance diversity and condition alignment. Our Condition-Annealed Diffusion Sampler (CADS) can be used with any pretrained model and sampling algorithm, and we show that it boosts the diversity of diffusion models in various conditional generation tasks. Further, using an existing pretrained diffusion model, CADS achieves a new state-of-the-art FID of 1.70 and 2.31 for class-conditional ImageNet generation at 256×256 and 512×512 respectively. DDPM CADS (Ours) DDPM CADS (Ours) DDPM CADS (Ours) DDPM CADS (Ours)

For agents

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

get_harvested_code_for_paper("2310.17347")
get_code_for_paper("2310.17347")
have("2310.17347")

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

curl https://syntology.ai/api/ran/2310.17347.json

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

[![Syntology run record](https://syntology.ai/api/ran/2310.17347.svg)](https://syntology.ai/paper/2310.17347)

Connect an agent — have() is free.