FiCo：面向语料库级电子表格问答的先找后算方法
原标题：FICO: Find-Then-Compute for Corpus-Level Spreadsheet Question Answering
AI 摘要
该论文提出 FiCo（Find-then-Compute）方法，用于跨电子表格语料库的问答任务：先检索文档摘要并消歧相似工作簿，再对选中完整表格执行受限 SQL 计算。在 DataBench（80 个数据集、1810 个问题）上达到 76.2% 准确率，比强 TableRAG 基线高 9.9 个百分点，比前缀 RAG 高 63.4 个百分点；在 MiMoTable 508 个问题上达 79.7%。研究还通过金标准工作簿干预揭示出 10.0 个百分点的来源选择成本，说明语义来源选择与精确计算结合是性能提升的关键。
正文节选
2nd Workshop on Agentic AI Benchmarks and Applications for Enterprise Tasks FiCo: Find-Then-Compute for Corpus-Level Spreadsheet Question Answering Abstract 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,