{"ID":2870316,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.12883","arxiv_id":"2509.12883","title":"Lego-Edit: A General Image Editing Framework with Model-Level Bricks and MLLM Builder","abstract":"Instruction-based image editing has garnered significant attention due to its direct interaction with users. However, real-world user instructions are immensely diverse, and existing methods often fail to generalize effectively to instructions outside their training domain, limiting their practical application. To address this, we propose Lego-Edit, which leverages the generalization capability of Multi-modal Large Language Model (MLLM) to organize a suite of model-level editing tools to tackle this challenge. Lego-Edit incorporates two key designs: (1) a model-level toolkit comprising diverse models efficiently trained on limited data and several image manipulation functions, enabling fine-grained composition of editing actions by the MLLM; and (2) a three-stage progressive reinforcement learning approach that uses feedback on unannotated, open-domain instructions to train the MLLM, equipping it with generalized reasoning capabilities for handling real-world instructions. Experiments demonstrate that Lego-Edit achieves state-of-the-art performance on GEdit-Bench and ImgBench. It exhibits robust reasoning capabilities for open-domain instructions and can utilize newly introduced editing tools without additional fine-tuning. Code is available: https://github.com/xiaomi-research/lego-edit.","short_abstract":"Instruction-based image editing has garnered significant attention due to its direct interaction with users. However, real-world user instructions are immensely diverse, and existing methods often fail to generalize effectively to instructions outside their training domain, limiting their practical application. To addr...","url_abs":"https://arxiv.org/abs/2509.12883","url_pdf":"https://arxiv.org/pdf/2509.12883v1","authors":"[\"Qifei Jia\",\"Yu Liu\",\"Yajie Chai\",\"Xintong Yao\",\"Qiming Lu\",\"Yasen Zhang\",\"Runyu Shi\",\"Ying Huang\",\"Guoquan Zhang\"]","published":"2025-09-16T09:36:17Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Reinforcement Learning\",\"Large Language Model\",\"Language Model\",\"Generative Adversarial Network\"]","has_code":false,"code_links":[{"ID":609756,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2870316,"paper_url":"https://arxiv.org/abs/2509.12883","paper_title":"Lego-Edit: A General Image Editing Framework with Model-Level Bricks and MLLM Builder","repo_url":"https://github.com/xiaomi-research/lego-edit","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}