arXiv 2605.074772026-05-08cs.CV

ReasonEdit：通过强化学习实现可解释图像编辑评估

ReasonEdit: Towards Interpretable Image Editing Evaluation via Reinforcement Learning

University of Electronic Science and Technology of China（电子科学与技术大学）；Shanghai Jiao Tong University（上海交通大学）

Honghua Chen, Zitong Xu, Huiyu Duan, Xinyun Zhang, Xiongkuo Min, Guangtao Zhai

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出ReasonEdit，通过引入ReasonEdit-22K数据集和RE-Reward模型，训练出可解释的图像编辑评估模型，提升评估的可解释性和透明度。

摘要：

近期文本引导图像编辑（TIE）模型取得了显著进展，但许多编辑结果仍存在伪影、意外修改和次优美学问题。尽管已有多个基准和评估方法，但大多数现有方法依赖标量评分且缺乏可解释性。这一限制主要源于缺乏高质量的TIE解释数据集和有效的奖励模型来训练可解释的评估器。为解决这些挑战，我们引入了ReasonEdit-22K，这是首个结合22K编辑图像和113K链式思考（CoT）样本的数据集，以及130万人类判断评估这些解释在逻辑性、准确性和有用性方面的表现。基于此数据集，我们提出了RE-Reward，一种基于多模态大语言模型（MLLM）的奖励模型，旨在为图像编辑的可解释推理评估提供人类对齐的反馈。此外，我们开发了ReasonEdit，该模型通过来自RE-Reward和群体相对策略优化（GRPO）算法的奖励信号进行训练，以学习可解释的评估模型。大量实验表明，ReasonEdit在与人类偏好的一致性方面表现优异，并在公共基准上表现出强大的泛化能力。此外，它能够生成高质量的可解释评估文本，使图像编辑的评估更加透明和可信。代码可在https://github.com/IntMeGroup/ReasonEdit获取。

英文摘要：

Recent text-guided image editing (TIE) models have achieved remarkable progress, however, many edited results still suffer from artifacts, unintended modifications, and suboptimal aesthetics. Although several benchmarks and evaluation methods have been proposed, most existing approaches rely on scalar scores and lack interpretability. This limitation largely stems from the absence of high-quality interpretation datasets for TIE and effective reward models to train interpretable evaluators. To address these challenges, we introduce ReasonEdit-22K, the first dataset that combines 22K edited images with 113K Chain-of-Thought (CoT) samples, along with 1.3M human judgments assessing these interpretations in terms of logicality, accuracy, and usefulness. Building upon this dataset, we propose RE-Reward, a multimodal large language model (MLLM)-based reward model designed to provide human-aligned feedback for evaluating interpretable reasoning in image editing. Furthermore, we develop ReasonEdit, which is trained using reward signals derived from RE-Reward and the Group Relative Policy Optimization (GRPO) algorithm to learn an interpretable evaluation model. Extensive experiments demonstrate that ReasonEdit achieves superior alignment with human preferences and exhibits strong generalization across public benchmarks. In addition, it is capable of generating high-quality interpretable evaluation text, enabling more transparent and trustworthy assessment for image editing. The code is available at https://github.com/IntMeGroup/ReasonEdit.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。