MTEdit：多轮图像编辑中，打包轨迹优于回滚
MTEdit: Packed Trajectories Beat Rollouts for Multi-Turn Image Editing
AI总结：
该研究提出多轮图像编辑数据集MTEdit及基准MTEdit-Bench，微调Qwen模型得到Qwen-Image-MTEdit，实验显示其在开源模型中性能最优，验证了MTEdit的有效性。
摘要：
多轮图像编辑旨在根据一系列指令修改图像，相比单轮编辑更适配人机协同创作。然而，构建高质量的多轮训练数据仍具挑战，因为传统的顺序回滚会在编辑轮次间累积误差并引入视觉质量漂移。为应对这一挑战，我们提出MTEdit，这是一个以打包轨迹画布（PTC）生成为核心流程构建的多轮图像编辑数据集。通过在统一视觉上下文内联合生成多个编辑状态，该流程减少了轮次间的视觉质量漂移。MTEdit涵盖分步编辑与上下文感知编辑场景，包含50万条多轮图像编辑轨迹，每条轨迹最多8轮，平均每轮5.38轮，是目前规模最大的多轮图像编辑数据集之一。我们还提出MTEdit-Bench，这是一个通过评估模型遵循编辑指令、保持内容一致性、抵抗视觉质量漂移及有效利用轮次间历史上下文的能力，来评估上述两种场景的基准。我们使用MTEdit对Qwen-Image-Edit-Causal进行微调，得到Qwen-Image-MTEdit。在MTEdit-Bench上，我们评估了包括Qwen-Image-MTEdit在内的大量闭源与开源图像编辑模型。实验结果显示，Qwen-Image-MTEdit在被评估的开源模型中取得了最佳性能，证明了MTEdit在提升模型多轮图像编辑能力方面的有效性。
英文摘要：
Multi-turn image editing aims to modify an image according to a sequence of instructions, making it better aligned with human-AI collaborative creation than single-turn editing. However, constructing high-quality multi-turn training data remains challenging, as conventional sequential rollout accumulates errors and introduces visual-quality drift across editing turns. To address this challenge, we introduce MTEdit, a multi-turn image editing dataset built through a pipeline centered on packed trajectory canvas (PTC) generation. By jointly generating multiple editing states within a unified visual context, this pipeline reduces visual-quality drift across turns. Covering both step-wise and context-aware editing scenarios, MTEdit comprises 500K multi-turn image editing trajectories, with up to 8 turns and an average of 5.38 turns per trajectory, making it one of the most extensive datasets for multi-turn image editing. We further introduce MTEdit-Bench, a benchmark designed to evaluate both scenarios by assessing models' ability to follow editing instructions, maintain content consistency, resist visual-quality drift, and effectively use historical context across turns. Using MTEdit, we fine-tune Qwen-Image-Edit-Causal to obtain Qwen-Image-MTEdit. On MTEdit-Bench, we evaluate a broad range of closed-source and open-source image editing models, including Qwen-Image-MTEdit. Experimental results show that Qwen-Image-MTEdit achieves the best performance among the evaluated open-source models, demonstrating the effectiveness of MTEdit in improving models' multi-turn image editing capabilities.
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。
