---
title: README
emoji: 🚀
colorFrom: blue
colorTo: green
sdk: static
pinned: false
---

<div style="display: flex; flex-direction: row; align-items: center; justify-content: center;"">
  <img src="https://cdn-uploads.huggingface.co/production/uploads/6320c537a023aad6a7680c8b/Ua2SoIM47E8cbntSg8F-7.png" alt="drawing" width="100" height="100" style="margin-right: 10px;"/>
</div>

Welcome to the DCPO (**D**ual **C**aption **P**reference **O**ptimization) organization! We're the authors of the Pick-Double Caption dataset. We have hosted the dataset, models
trained on the dataset, and other artifacts under this organization. You will be able to find them below. Additionally, here are the links
to the paper and the project website:

* [arXiv](https://arxiv.org/abs/2502.06023v2)
* [Paper](https://openreview.net/pdf?id=ruZksIJBBd)
* [Codebase](https://github.com/sahsaeedi/DCPO-T2I)


  **Update October 23, 2025:** We got accepted to TMLR journal 🥳