CRAFT：无训练 cascaded 检索用于表格问答
CRAFT: Training-Free Cascaded Retrieval for Tabular QA
AI总结：
CRAFT 提出了一种无训练 cascaded 检索方法，通过稀疏检索模型筛选候选表格后，再用密集模型进行重排序，提升检索质量并在 NQ-Tables 和 OTT-QA 数据集上表现优异。
摘要：
开放领域表格问答（TQA）涉及从大规模语料库中检索相关表格以回答自然语言查询。传统密集检索模型如 DTR 和 DPR 在大规模检索任务中计算成本高且需重新训练或微调新数据集，限制了其在演变领域和知识中的适应性。我们提出 CRAFT，一种零样本 cascaded 检索方法，首先使用稀疏检索模型筛选候选表格子集，再应用更计算密集的密集模型作为重排序器。为提高检索质量，我们通过 Gemini Flash 1.5 生成描述性标题和摘要来丰富表格表示，使查询与表格结构之间的语义匹配更丰富。我们的方法在 NQ-Tables 数据集上优于最先进的稀疏、密集和混合检索器，并在更具挑战性的 OTT-QA 评估中表现出色，达到更高的召回阈值，其中任务要求跨文本段落和关系表格进行多跳推理。本工作建立了一种可扩展且适应性强的表格检索范式，弥合了微调架构与轻量级、即插即用检索系统之间的差距。代码和数据可在 https://coral-lab-asu.github.io/CRAFT/ 获取。
英文摘要：
Open-Domain Table Question Answering (TQA) involves retrieving relevant tables from a large corpus to answer natural language queries. Traditional dense retrieval models such as DTR and DPR incur high computational costs for large-scale retrieval tasks and require retraining or fine-tuning on new datasets, limiting their adaptability to evolving domains and knowledge. We propose CRAFT, a zero-shot cascaded retrieval approach that first uses a sparse retrieval model to filter a subset of candidate tables before applying more computationally expensive dense models as re-rankers. To improve retrieval quality, we enrich table representations with descriptive titles and summaries generated by Gemini Flash 1.5, enabling richer semantic matching between queries and tabular structures. Our method outperforms state-of-the-art sparse, dense, and hybrid retrievers on the NQ-Tables dataset. It also demonstrates strong zero-shot performance on the more challenging OTT-QA benchmark, achieving competitive results at higher recall thresholds, where the task requires multi-hop reasoning across both textual passages and relational tables. This work establishes a scalable and adaptable paradigm for table retrieval, bridging the gap between fine-tuned architectures and lightweight, plug-and-play retrieval systems. Code and data are available at https://coral-lab-asu.github.io/CRAFT/
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。
