arXiv 2603.206442026-03-24cs.CV

ScaleEdit-12M：通过多智能体框架实现开源图像编辑数据生成的规模化

ScaleEdit-12M: Scaling Open-Source Image Editing Data Generation via Multi-Agent Framework

Shanghai Jiao Tong University（上海交通大学）；Shanghai AI Laboratory（上海人工智能实验室）；CUHK MMLab（香港中文大学多模态实验室）；Fudan University（复旦大学）；University of Science and Technology of China（中国科学技术大学）；Xiamen University（厦门大学）

Guanzhou Chen, Erfei Cui, Changyao Tian, Danni Yang, Ganlin Yang, Yu Qiao, Hongsheng Li, Gen Luo, Hongjie Zhang

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出ScaleEditor框架，通过多智能体方法构建大规模高质量图像编辑数据集，生成ScaleEdit-12M，涵盖23类任务，提升多种编辑模型性能。

摘要：

基于指令的图像编辑已成为统一多模态模型（UMMs）的关键能力，但构建大规模、多样化且高质量的编辑数据集仍具挑战性。此前的图像编辑数据集要么依赖闭源模型标注，限制了成本效益的扩展；要么采用固定合成编辑流程，导致质量和泛化能力有限。为解决这些问题，我们提出ScaleEditor，一个完全开源的分层多智能体框架，用于端到端构建大规模高质量图像编辑数据集。我们的流程包括三个关键组件：源图像扩展与世界知识注入、自适应多智能体编辑指令-图像合成、以及任务感知的数据质量验证机制。使用ScaleEditor，我们整理出ScaleEdit-12M，目前最大的开源图像编辑数据集，覆盖23类任务，涵盖多样化的现实和合成领域。在ScaleEdit上微调UniWorld-V1和Bagel，取得了显著提升，包括在ImgEdit上提升10.4%，在GEdit上提升35.1%，在知识注入基准上分别达到150.0%和26.5%。这些结果表明，开源、智能体流程可以达到商业级数据质量，同时保持成本效益和可扩展性。框架和数据集将开源发布。

英文摘要：

Instruction-based image editing has emerged as a key capability for unified multimodal models (UMMs), yet constructing large-scale, diverse, and high-quality editing datasets without costly proprietary APIs remains challenging. Previous image editing datasets either rely on closed-source models for annotation, which prevents cost-effective scaling, or employ fixed synthetic editing pipelines, which suffer from limited quality and generalizability. To address these challenges, we propose ScaleEditor, a fully open-source hierarchical multi-agent framework for end-to-end construction of large-scale, high-quality image editing datasets. Our pipeline consists of three key components: source image expansion with world-knowledge infusion, adaptive multi-agent editing instruction-image synthesis, and a task-aware data quality verification mechanism. Using ScaleEditor, we curate ScaleEdit-12M, the largest open-source image editing dataset to date, spanning 23 task families across diverse real and synthetic domains. Fine-tuning UniWorld-V1 and Bagel on ScaleEdit yields consistent gains, improving performance by up to 10.4% on ImgEdit and 35.1% on GEdit for general editing benchmarks and by up to 150.0% on RISE and 26.5% on KRIS-Bench for knowledge-infused benchmarks. These results demonstrate that open-source, agentic pipelines can approach commercial-grade data quality while retaining cost-effectiveness and scalability. Both the framework and dataset will be open-sourced.

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。