Pico-Banana-400K：一个用于文本引导图像编辑的大规模数据集
Pico-Banana-400K: A Large-Scale Dataset for Text-Guided Image Editing
该数据集为文本引导图像编辑模型提供了大规模、高质量的训练与评测资源，特别是其多轮编辑、偏好对齐和指令重写子集，可直接用于研究编辑的序列规划、奖励模型训练和指令理解，推动更鲁棒的编辑系统发展。
原文依据 We introduce Pico-Banana-400K, a comprehensive 400K-image dataset for instruction-based image editing.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论近期多模态模型的进展展示了卓越的文本引导图像编辑能力，GPT-4o 和 Nano-Banana 等系统设立了新的基准。然而，研究社区的进展仍受限于缺乏基于真实图像构建的大规模、高质量且公开可用的数据集。我们推出 Pico-Banana-400K，一个全面的 40 万图像数据集，用于基于指令的图像编辑。我们的数据集利用 Nano-Banana 从 OpenImages 集合中的真实照片生成多样化的编辑对。Pico-Banana-400K 与以往合成数据集的不同之处在于我们系统化的质量和多样性方法。我们采用细粒度图像编辑分类法，确保全面覆盖编辑类型，同时通过基于多模态大语言模型的质量评分和精心整理来保持精确的内容保留和指令忠实度。除了单轮编辑，Pico-Banana-400K 还支持复杂编辑场景的研究。该数据集包括三个专门子集：（1）一个 7.2 万示例的多轮集合，用于研究连续修改中的顺序编辑、推理和规划；（2）一个 5.6 万示例的偏好子集，用于对齐研究和奖励模型训练；（3）配对的长短编辑指令，用于开发指令重写和摘要能力。通过提供这一大规模、高质量且任务丰富的资源，Pico-Banana-400K 为训练和基准测试下一代文本引导图像编辑模型奠定了坚实基础。
英文原文摘要 1602 字符
Recent advances in multimodal models have demonstrated remarkable text-guided image editing capabilities, with systems like GPT-4o and Nano-Banana setting new benchmarks. However, the research community's progress remains constrained by the absence of large-scale, high-quality, and openly accessible datasets built from real images. We introduce Pico-Banana-400K, a comprehensive 400K-image dataset for instruction-based image editing. Our dataset is constructed by leveraging Nano-Banana to generate diverse edit pairs from real photographs in the OpenImages collection. What distinguishes Pico-Banana-400K from previous synthetic datasets is our systematic approach to quality and diversity. We employ a fine-grained image editing taxonomy to ensure comprehensive coverage of edit types while maintaining precise content preservation and instruction faithfulness through MLLM-based quality scoring and careful curation. Beyond single turn editing, Pico-Banana-400K enables research into complex editing scenarios. The dataset includes three specialized subsets: (1) a 72K-example multi-turn collection for studying sequential editing, reasoning, and planning across consecutive modifications; (2) a 56K-example preference subset for alignment research and reward model training; and (3) paired long-short editing instructions for developing instruction rewriting and summarization capabilities. By providing this large-scale, high-quality, and task-rich resource, Pico-Banana-400K establishes a robust foundation for training and benchmarking the next generation of text-guided image editing models.