PARTAB：面向可扩展表格理解的、结合结构化证据的分区感知推理
PARTAB: Partition-Aware Reasoning with Structured Evidence for Scalable Table Understanding
- University of Alberta（阿尔伯塔大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本研究提出PARTAB框架，通过构建结构化证据接口、分层选择表格证据，提升LLM在大规模表格推理任务中的性能，在多个基准上表现优于现有方法。
AI中文摘要：
大型语言模型（LLM）在表格推理方面展现出强大能力，但随着表格规模增大和复杂度提升，因存在无关上下文且难以定位推理所需证据，其有效性会下降。现有方法通常对完整表格或单一简化视图进行推理，这仍可能掩盖重要的行-列关系。我们提出PARTAB（Partition-Aware Reasoning over Tables，即面向表格的分区感知推理），这是一种在LLM与表格间构建结构化证据接口的框架。PARTAB将与查询相关的证据表示为语义连贯、行关联的表格区域，在列组和行级分区上执行分层选择，之后组合所选证据以生成答案。我们在多个表格推理基准上对PARTAB进行评估，涵盖问答、事实验证和数值推理任务。PARTAB的表现始终优于完整表格提示及多种近期表格推理方法，在WikiTableQuestions和TabFact上取得优异性能，同时在数值推理任务上保持竞争力。额外分析显示，语义分区和针对性证据选择可提升证据定位能力，大幅减少推理上下文，并在复杂表格上带来更大益处。这些结果表明，结构化、分区感知的证据构建对可扩展表格推理具有重要价值。
英文摘要：
Large Language Models (LLMs) have shown strong capabilities in table reasoning, but their effectiveness degrades as tables grow in size and complexity due to irrelevant context and difficulty localizing the evidence required for reasoning. Existing approaches typically reason over either the full table or a single reduced view, which can still obscure important row-column relationships. We introducePARTAB (Partition-Aware Reasoning overTables), a framework that constructs a structured evidence interface between the LLM and the table. PARTAB represents query-relevant evidence as semantically coherent, row-linked table regions and performs hierarchical selection over column groups and row-level partitions before composing the selected evidence for answer generation. We evaluate PARTAB on multiple table reasoning benchmarks, covering question answering, fact verification, and numerical reasoning. PARTAB consistently improves over full-table prompting and several recent table reasoning methods, achieving strong performance on WikiTableQuestions and TabFact while remaining competitive on numerical reasoning. Additional analyses show that semantic partitioning and targeted evidence selection improve evidence localization, substantially reduce the reasoning context, and provide larger benefits on complex tables. These results demonstrate the value of structured, partition aware evidence construction for scalable table reasoning.
