发表机构
Carnegie Mellon University; Atlassian（卡内基梅隆大学; 阿特lassian）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对电子表格集合问答，提出先查找后计算（FiCo）方法，结合文档检索、工作簿消歧与SQL执行，在DataBench和MiMoTable上显著优于基线，并揭示源选择成本。
AI 中文摘要
电子表格集合上的问答需要找到正确的工作簿，并对完整表格进行计算。我们提出了“先查找后计算”（Find-then-Compute，FiCo）方法，该方法检索文档摘要、消除相似工作簿的歧义，并在选定的完整表格上执行受约束的结构化查询语言（SQL）。在DataBench（80个数据集，1810个问题）上，FiCo达到了76.2%的准确率：在轨道预指定的评估器下，比使用相同冻结工作簿选择的强TableRAG风格基线（66.3%）高出9.9个百分点，比前缀RAG（12.8%）高出63.4个百分点。在508个MiMoTable问题上，FiCo达到79.7%，而前缀RAG仅为22.2%。将黄金工作簿提供给强基线，使其从66.3%提升至76.3%，揭示了在固定计算下10.0个百分点的源选择成本。尽管文档召回率达到95.1%，可执行SQL达到98.5%，但只有81.3%的问题在黄金工作簿上成功执行。FiCo的优势在于将语义源选择与精确的、基于模式的計算相整合。
英文摘要
Question answering over spreadsheet collections requires finding the correct workbook and computing over complete tables. We introduce Find-then-Compute (FiCo), which retrieves document summaries, disambiguates similar workbooks, and executes constrained Structured Query Language (SQL) over the selected full table. On DataBench (80 datasets, 1,810 questions), FiCo reaches 76.2% accuracy: 9.9 points above a strong TableRAG-style baseline on the same frozen workbook choices (66.3%) under the tracks' prespecified evaluators, and 63.4 points above prefix RAG (12.8%). On 508 MiMoTable questions, FiCo reaches 79.7%, versus 22.2% for prefix RAG. Giving the strong baseline the gold workbook raises it from 66.3% to 76.3%, exposing a 10.0-point source-selection cost under fixed compute. Despite 95.1% document recall and 98.5% executable SQL, only 81.3% of questions execute on the gold workbook. FiCo's advantage comes from integrating semantic source selection with exact, schema-grounded computation.
