UnicEdit-10M: 一个通过统一验证打破规模-质量壁垒的数据集和基准
UnicEdit-10M: A Dataset and Benchmark Breaking the Scale-Quality Barrier via Unified Verification for Reasoning-Enriched Edits
AI总结:
UnicEdit-10M通过统一验证方法构建大规模高质量数据集和基准,解决图像编辑中规模与质量的矛盾,评估模型在空间和知识推理中的表现。
摘要:
随着强大多模态模型如GPT-4o、Nano Banana和Seedream 4.0在图像编辑领域的快速发展,闭源模型与开源模型之间的性能差距正在扩大,主要原因是大规模、高质量训练数据和能够诊断模型在多样化编辑行为中弱点的综合基准的稀缺。现有数据构建方法面临规模与质量的权衡:人工标注高质量但不可扩展,而自动化流程则容易产生错误传播和噪声。为此,我们引入了一个轻量级的数据流水线,用端到端模型和统一的后验证阶段取代多工具链。为了实现可扩展的质量控制,我们训练了一个7B双任务专家模型,即Qwen-Verify,用于高效的失败检测和指令重述。该流水线产生了UnicEdit-10M数据集,涵盖多样化的基础和复杂编辑任务。我们还提出了UnicBench通用基准,其范围超出了基础编辑,明确评估空间和知识驱动的推理。为了实现细粒度诊断,我们引入了新的指标,包括非编辑一致性与推理准确性。我们对主流模型在UnicBench上的分析揭示了它们的局限性,并为未来研究提供了清晰的方向。
英文摘要:
With the rapid advances of powerful multimodal models such as GPT-4o, Nano Banana, and Seedream 4.0 in Image Editing, the performance gap between closed-source and open-source models is widening, primarily due to the scarcity of large-scale, high-quality training data and comprehensive benchmarks capable of diagnosing model weaknesses across diverse editing behaviors. Existing data construction methods face a scale-quality trade-off: human annotations are high-quality but not scalable, while automated pipelines suffer from error propagation and noise. To address this, we introduce a lightweight data pipeline that replaces multi-toolchains with an end-to-end model and a unified post-verification stage. For scalable quality control, we train a 7B dual-task expert model, \textbf{Qwen-Verify}, for efficient failure detection and instruction recaptioning. This pipeline yields \textbf{UnicEdit-10M}, a 10M-scale dataset spanning diverse basic and complex editing tasks. We also propose \textbf{UnicBench}, a general benchmark that extends beyond basic edits to explicitly assess spatial and knowledge-driven reasoning. To enable fine-grained diagnosis, we introduce novel metrics, including \textit{Non-edit Consistency} and \textit{Reasoning Accuracy}. Our analysis of mainstream models on UnicBench reveals their limitations and provides clear directions for future research.
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。