用图结构显式规划推理路径，提升复杂表格问答准确率
GraphOTTER: Evolving LLM-based Graph Reasoning for Complex Table Question Answering
- 将复杂表格转为无向图，分步执行预设推理动作
- 在两个数据集上优于现有方法，尤其擅长过滤无关信息
- 适合需要可解释推理的表格问答场景
复杂表格问答要求基于布局复杂、表头位置灵活的表格提供准确答案。尽管大语言模型时代已取得显著进展，但现有方法的推理过程常不透明，直接将整张表输入提示词，难以有效剔除无关信息。为此，我们提出GraphOTTER，显式构建推理流程以精准定位答案。具体而言，GraphOTTER采用图表示法，将复杂表格转换为无向图，并在此基础上分步推理，每一步由一组预定义的中间推理动作引导。由此生成清晰的推理路径，有效识别问题答案。在两个基准数据集和两种LLM骨干模型上的全面实验表明，GraphOTTER表现优异。进一步分析显示，其成功可能源于高效过滤无关信息的能力，使推理聚焦于关键数据。代码与实验数据集见：https://github.com/JDing0521/GraphOTTER。
原文摘要 · Abstract (English)
Complex Table Question Answering involves providing accurate answers to specific questions based on intricate tables that exhibit complex layouts and flexible header locations. Despite considerable progress having been made in the LLM era, the reasoning processes of existing methods are often implicit, feeding the entire table into prompts, making it difficult to effectively filter out irrelevant information in the table. To this end, we propose GraphOTTER that explicitly establishes the reasoning process to pinpoint the correct answers. In particular, GraphOTTER leverages a graph-based representation, transforming the complex table into an undirected graph. It then conducts step-by-step reasoning on the graph, with each step guided by a set of pre-defined intermediate reasoning actions. As such, it constructs a clear reasoning path and effectively identifies the answer to a given question. Comprehensive experiments on two benchmark datasets and two LLM backbones demonstrate the effectiveness of GraphOTTER. Further analysis indicates that its success may be attributed to the ability to efficiently filter out irrelevant information, thereby focusing the reasoning process on the most pertinent data. Our code and experimental datasets are available at \url{https://github.com/JDing0521/GraphOTTER}.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。