English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We present Direct Reward Fine-Tuning (DRaFT), a simple and effective methodfor fine-tuning diffusion models to maximize differentiable reward functions,such as scores from human preference models. We first show that it is possibleto backpropagate the reward function gradient through the full samplingprocedure, and that doing so achieves strong performance on a variety ofrewards, outperforming reinforcement learning-based approaches. We then proposemore efficient variants of DRaFT: DRaFT-K, which truncates backpropagation toonly the last K steps of sampling, and DRaFT-LV, which obtains lower-variancegradient estimates for the case when K=1. We show that our methods work wellfor a variety of reward functions and can be used to substantially improve theaesthetic quality of images generated by Stable Diffusion 1.4. Finally, we drawconnections between our approach and prior work, providing a unifyingperspective on the design space of gradient-based fine-tuning algorithms.

Directly Fine-Tuning Diffusion Models on Differentiable Rewards