发表机构
Harbin Institute of Technology; AI Research Center, Midea Group (Shanghai) Co., Ltd.; Changchun University of Science and Technology（哈尔滨工业大学; 美的集团（上海）有限公司AI研究中心; 长春理工大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对大语言模型处理表格时忽略其二维与分层结构的问题，提出H2Table框架，通过分层嵌套超图及定制超图编码器等，在HiTab数据集嵌套深度4的复杂表格任务上较最优基线提升22.88%。
AI 中文摘要
表格在不同领域中普遍存在，但对其进行推理仍是现代大语言模型（LLMs）面临的重大挑战。当前方法通常将表格线性化为序列，固有地忽略了其内在的二维和分层结构。为解决这一问题，我们提出H2Table（分层超图增强型表格推理），这是一个将复杂表格表示为分层嵌套超图的新型框架。为处理该表示，我们设计了定制的超图编码器，以促进超边（表头）与节点（单元格）之间的消息传递，从而感知复杂表格内它们之间的语义蕴含关系。此外，我们引入一组可学习的查询向量，作为轻量型桥梁，将编码器中的代表性结构嵌入提取至LLM中。实验结果表明，我们的方法能有效处理具有分层嵌套表头的复杂表格问答任务。值得注意的是，在HiTab数据集上，H2Table在嵌套深度为4的高度复杂表格上，较现有最优基线实现了22.88%的平均性能提升。我们的代码可在以下URL获取：this https URL。
英文摘要
Tables are ubiquitous across diverse domains, yet reasoning over them remains a significant challenge for modern large language models (LLMs). Current approaches typically linearize tables into sequences, inherently overlooking their intrinsic two-dimensional and hierarchical structure. To address this, we propose H2Table (Hierarchical Hypergraph-Enhanced Table Reasoning), a novel framework that represents complex tables as hierarchical nested hypergraphs. To process this representation, we design a tailored hypergraph encoder to facilitate message passing between hyperedges (headers) and nodes (cells), thereby perceiving the semantic entailment relationships between them within complex tables. Furthermore, we introduce a set of learnable query vectors acting as a lightweight bridge to extract representative structural embeddings from the encoder into the LLM. Experimental results demonstrate that our approach effectively handles complex table question answering tasks with hierarchical nested headers. Notably, on the HiTab dataset, H2Table achieves an average improvement of 22.88% over state-of-the-art baselines on highly complex tables with a nesting depth of four. Our code is available at: https://github.com/lila120/h2table.
