ARK：一个双轴多模态检索基准，沿推理与知识
ARK: A Dual-Axis Multimodal Retrieval Benchmark along Reasoning and Knowledge
AI总结：
ARK基准通过双轴视角评估多模态检索，揭示知识密集型与推理密集型检索间的差距，发现细粒度视觉推理是主要瓶颈。
摘要：
现有多模态检索基准大多强调日常图像上的语义匹配，并对专业知识和复杂推理提供有限的诊断。为解决这一差距，我们引入ARK，一个旨在从两个互补的视角分析多模态检索的基准：(i) 知识领域（五个领域，17个子类型），这定义了检索所依赖的内容和专业知识；(ii) 推理技能（六个类别），这定义了在多模态证据上所需的推理类型以确定正确候选。具体而言，ARK评估了单模态和多模态查询和候选的检索，涵盖了16种异质视觉数据类型。为了避免评估中的捷径匹配，大多数查询都与需要多步骤推理的目标困难负样本配对。我们在ARK上评估了23种代表性的文本和多模态检索器，并观察到知识密集型和推理密集型检索之间存在显著差距，细粒度视觉和空间推理成为持续的瓶颈。我们进一步表明，简单的增强如重排序和重写会产生一致的改进，但仍有大量提升空间。
英文摘要：
Existing multimodal retrieval benchmarks largely emphasize semantic matching on daily-life images and offer limited diagnostics of professional knowledge and complex reasoning. To address this gap, we introduce ARK, a benchmark designed to analyze multimodal retrieval from two complementary perspectives: (i) knowledge domains (five domains with 17 subtypes), which characterize the content and expertise retrieval relies on, and (ii) reasoning skills (six categories), which characterize the type of inference over multimodal evidence required to identify the correct candidate. Specifically, ARK evaluates retrieval with both unimodal and multimodal queries and candidates, covering 16 heterogeneous visual data types. To avoid shortcut matching during evaluation, most queries are paired with targeted hard negatives that require multi-step reasoning. We evaluate 25 representative text-based and multimodal retrievers and observe a pronounced gap between knowledge- and reasoning-intensive retrieval, with fine-grained visual and spatial reasoning as persistent bottlenecks. We further show that enhancements such as re-ranking, rewriting, and agentic retrieval yield consistent gains, but substantial headroom remains.
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。
