meituan-longcat/CEdit-Bench
收藏资源简介:
--- license: mit task_categories: - image-to-image language: - en - zh pretty_name: CEdit-Bench size_categories: - 1K<n<10K --- **CEdit-Bench** is a comprehensive evaluation suite, first proposed in the **LongCat-Image technical report**, and developed by integrating and extending existing image editing benchmarks. We further curate new data to enhance task diversity, yielding a robust dataset of 1,464 bilingual (Chinese–English) editing pairs across 15 fine-grained task categories, providing a more holistic and rigorous standard for evaluating image editing models. <div align="center"> <img src="assets/data_distribution.jpg" width="75%" alt="data distribution." /> </div> An example entry is shown below: ``` { "key": "730fc5ea-7c14-45c9-a2a8-8d802b49a610", "input_image": bytes, "instruction_en": "Adjust the background to a garden", "instruction_cn": "将背景调整为花园", "task_type": "Background Change" } ```
许可证:MIT协议 task_categories: - 图像到图像(image-to-image) language: - 英语 - 中文 pretty_name: CEdit-Bench size_categories: - 1000 < 样本数 < 10000 **CEdit-Bench**是一款综合性图像编辑评测套件,首次提出于《LongCat-Image技术报告》,通过整合并拓展现有图像编辑基准数据集开发而成。我们进一步精选新增数据以丰富任务多样性,最终构建出涵盖15个细粒度任务类别、共计1464组中英双语编辑样本对的高质量数据集,为图像编辑模型的评测提供了更全面且严谨的标准。 <div align="center"> <img src="assets/data_distribution.jpg" width="75%" alt="数据集分布" /> </div> 以下为一条示例条目: { "唯一标识": "730fc5ea-7c14-45c9-a2a8-8d802b49a610", "输入图像": 二进制字节流, "英文指令": "Adjust the background to a garden", "中文指令": "将背景调整为花园", "任务类型": "背景修改" }




