发表机构
RMIT University; University of Wollongong; The University of Queensland; Xiamen University（皇家墨尔本理工大学; 伍伦贡大学; 昆士兰大学; 厦门大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出SemiBonsai框架，通过表格结构化、不确定性消解、计划推理和预算感知路由，在三个基准上同时提升半结构化表格数值问答的准确率与成本效益。
AI 中文摘要
半结构化表格通过多样的布局元素（如分层行标题和列标题）编码了丰富的语义信息。对此类数据回答数值问题具有挑战性，因为它需要联合进行表格的准确结构理解、问题不确定性消解以及复杂查询意图的解读。现有方法很少能整体性地有效处理上述多方面挑战；此外，它们依赖大语言模型（LLM）而未考虑财务影响。我们提出SemiBonsai，一个经济高效、由LLM驱动的框架，其特点包括：（1）多路分层表格结构化器（Multiway Layered Table Structurer），将表格转换为按结构元素组织的多路分层树；（2）上下文感知的不确定性消解器（Context-Aware Uncertainty Resolver），将未明确指代的短语锚定到上下文一致的结构元素；（3）计划引导的推理器（Plan-Guided Reasoner），将复杂查询意图分解为查询计划以进行推理；（4）预算感知的强盗路由器（Budget-Aware Bandit Router），学习每个实例的LLM效用并在固定货币约束下优化LLM选择。在三个基准上的实验表明，SemiBonsai同时提升了问答有效性和成本效益。值得注意的是，表格结构化器相比替代树构建方法将答案准确率提高了46%，而路由器在各类预算下相比现有路由方法在答案准确率上实现了平均5%的相对提升。
英文摘要
Semi-structured tables encode rich semantic information through diverse layout elements, such as hierarchical row and column headers. Answering numerical questions over such data is challenging because it requires a joint effort of accurate structural understanding of tables, question uncertainty resolution, and complex query intents interpretation. Existing methods are rarely effective in handling the above multifaceted challenges in a holistic manner; moreover, they rely on LLMs without considering financial implications. We propose SemiBonsai, a cost-effective, LLM-powered framework featuring: (1) a Multiway Layered Table Structurer that converts a table into a multiway layered tree organized by structural elements; (2) a Context-Aware Uncertainty Resolver that grounds underspecified phrases to context-coherent structural elements; (3) a Plan-Guided Reasoner that decomposes complex query intents into a query plan for reasoning; (4) a Budget-Aware Bandit Router that learns per-instance LLM utilities and optimizes LLM selection under fixed monetary constraints. Experiments on three benchmarks show that SemiBonsai improves both QA effectiveness and cost-effectiveness. Notably, the Table Structurer improves answer accuracy by 46% over alternative tree construction methods, and Router achieves a 5% average relative gain in answer accuracy compared with existing routing methods across budgets.
CommentsThis is the technical report for the ICDE 2027 paper
