EditHF-1M: A Million-Scale Rich Human Preference Feedback for Image Editing
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914396227239936 |
|---|---|
| author | Xu, Zitong Duan, Huiyu Ji, Zhongpeng Zhang, Xinyun Liu, Yutao Min, Xiongkuo Gu, Ke Zhang, Jian Xu, Shusong Chen, Jinwei Li, Bo Zhai, Guangtao |
| author_facet | Xu, Zitong Duan, Huiyu Ji, Zhongpeng Zhang, Xinyun Liu, Yutao Min, Xiongkuo Gu, Ke Zhang, Jian Xu, Shusong Chen, Jinwei Li, Bo Zhai, Guangtao |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_14916 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | EditHF-1M: A Million-Scale Rich Human Preference Feedback for Image Editing Xu, Zitong Duan, Huiyu Ji, Zhongpeng Zhang, Xinyun Liu, Yutao Min, Xiongkuo Gu, Ke Zhang, Jian Xu, Shusong Chen, Jinwei Li, Bo Zhai, Guangtao Computer Vision and Pattern Recognition Multimedia 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. |
| title | EditHF-1M: A Million-Scale Rich Human Preference Feedback for Image Editing |
| topic | Computer Vision and Pattern Recognition Multimedia |
| url | https://arxiv.org/abs/2603.14916 |