SPARTA：可扩展且原则性的树状多跳问答基准
SPARTA: Scalable and Principled Benchmark of Tree-Structured Multi-hop QA over Text and Tables
AI总结：
SPARTA通过自动生成大规模表格-文本问答基准测试，揭示了跨模态推理的深层缺陷。
摘要：
现实中的表格-文本问答任务需要能够跨长文本和源表格推理的模型，需要跨多个跳转并执行复杂的操作，如聚合。然而，现有的基准测试较小，手动整理的，因此容易出错，并且包含浅层问题，很少需要超过两个跳转或调用聚合、分组或其他可以自然语言查询表达的高级分析操作。我们提出了SPARTA，一个端到端的构建框架，能够自动生成大规模的表格-文本问答基准测试，通过轻量级的人工验证，只需要HybridQA的一半注释时间。该框架首先通过丰富每个源表格，添加其地面表格，其中元组是自动从随附的非结构化段落中提取的原子事实，构建参考事实数据库，然后合成嵌套查询，其嵌套谓词的数量与期望的跳转次数匹配。为确保每个SQL语句可执行，并且其语言表达产生流畅、自然的疑问句，我们提出了两种新的技术：基于溯源的细化，它重写任何语法有效的查询，其结果非空；以及现实结构强制，它限制生成到查询图的后序遍历。最终的流水线生成成千上万的高质量问题-答案对，涵盖在文本和表格之间的聚合、分组和深度多跳推理。在SPARTA上，达到HybridQA 70分以上或OTT-QA 50分以上的最先进模型会下降超过30个F1点，暴露了当前跨模态推理的根本缺陷。我们的基准测试、构建代码和基线模型可在https://github.com/pshlego/SPARTA/tree/main上获得。
英文摘要：
Real-world Table-Text question answering (QA) tasks require models that can reason across long text and source tables, traversing multiple hops and executing complex operations such as aggregation. Yet existing benchmarks are small, manually curated - and therefore error-prone - and contain shallow questions that seldom demand more than two hops or invoke aggregations, grouping, or other advanced analytical operations expressible in natural-language queries. We present SPARTA, an end-to-end construction framework that automatically generates large-scale Table-Text QA benchmarks with lightweight human validation, requiring only one quarter of the annotation time of HybridQA. The framework first constructs a reference fact database by enriching each source table with grounding tables whose tuples are atomic facts automatically extracted from the accompanying unstructured passages, then synthesizes nested queries whose number of nested predicates matches the desired hop count. To ensure that every SQL statement is executable and that its verbalization yields a fluent, human-sounding question, we propose two novel techniques: provenance-based refinement, which rewrites any syntactically valid query that returns a non-empty result, and realistic-structure enforcement, which confines generation to post-order traversals of the query graph. The resulting pipeline produces thousands of high-fidelity question-answer pairs covering aggregations, grouping, and deep multi-hop reasoning across text and tables. On SPARTA, state-of-the-art models that reach over 70 F1 on HybridQA or over 50 F1 on OTT-QA drop by more than 30 F1 points, exposing fundamental weaknesses in current cross-modal reasoning. Our benchmark, construction code, and baseline models are available at https://github.com/pshlego/SPARTA/tree/main.
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。
