Recent text-guided image editing (TIE) models have achieved remarkable progress, while many edited images still suffer from issues such as artifacts, unexpected editings, unaesthetic contents. Although some benchmarks and methods have been proposed for evaluating edited images, scalable evaluation models are still lacking, which limits the development of human feedback reward models for image editing. To address the challenges, we first introduce \textbf{EditHF-1M}, a million-scale image editing dataset with over 29M human preference pairs and 148K human mean opinion ratings, both evaluated from three dimensions, \textit{i.e.}, visual quality, instruction alignment, and attribute preservation. Based on EditHF-1M, we propose \textbf{EditHF}, a multimodal large language model (MLLM) based evaluation model, to provide human-aligned feedback from image editing. Finally, we introduce \textbf{EditHF-Reward}, which utilizes EditHF as the reward signal to optimize the text-guided image editing models through reinforcement learning. Extensive experiments show that EditHF achieves superior alignment with human preferences and demonstrates strong generalization on other datasets. Furthermore, we fine-tune the Qwen-Image-Edit using EditHF-Reward, achieving significant performance improvements, which demonstrates the ability of EditHF to serve as a reward model to scale-up the image editing. Both the dataset and code will be released in our GitHub repository: https://github.com/IntMeGroup/EditHF.

翻译：近年来，文本引导的图像编辑模型取得了显著进展，但许多编辑后的图像仍存在伪影、意外编辑、内容不美观等问题。尽管已有一些基准和方法被提出用于评估编辑后的图像，但可扩展的评估模型仍然缺乏，这限制了面向图像编辑的人类反馈奖励模型的发展。为应对这些挑战，我们首先引入了 **EditHF-1M**，这是一个百万规模的图像编辑数据集，包含超过 2900 万个人类偏好对和 14.8 万个人类平均意见评分，两者均从三个维度进行评估，即视觉质量、指令对齐和属性保持。基于 EditHF-1M，我们提出了 **EditHF**，一个基于多模态大语言模型的评估模型，用于提供与人类对齐的图像编辑反馈。最后，我们引入了 **EditHF-Reward**，它利用 EditHF 作为奖励信号，通过强化学习来优化文本引导的图像编辑模型。大量实验表明，EditHF 在人类偏好对齐方面表现出色，并在其他数据集上展现出强大的泛化能力。此外，我们使用 EditHF-Reward 对 Qwen-Image-Edit 进行微调，实现了显著的性能提升，这证明了 EditHF 作为奖励模型以规模化提升图像编辑的能力。数据集和代码均将在我们的 GitHub 仓库中发布：https://github.com/IntMeGroup/EditHF。