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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Learning from human feedback has been shown to improve text-to-image models.These techniques first learn a reward function that captures what humans careabout in the task and then improve the models based on the learned rewardfunction. Even though relatively simple approaches (e.g., rejection samplingbased on reward scores) have been investigated, fine-tuning text-to-imagemodels with the reward function remains challenging. In this work, we proposeusing online reinforcement learning (RL) to fine-tune text-to-image models. Wefocus on diffusion models, defining the fine-tuning task as an RL problem, andupdating the pre-trained text-to-image diffusion models using policy gradientto maximize the feedback-trained reward. Our approach, coined DPOK, integratespolicy optimization with KL regularization. We conduct an analysis of KLregularization for both RL fine-tuning and supervised fine-tuning. In ourexperiments, we show that DPOK is generally superior to supervised fine-tuningwith respect to both image-text alignment and image quality. Our code isavailable athttps://github.com/google-research/google-research/tree/master/dpok.

DPOK: Reinforcement Learning for Fine-tuning Text-to-Image Diffusion Models