用混合推理提升大模型对表格的理解能力
TableMaster: A Recipe to Advance Table Understanding with Language Models
- 提取表格内容并注入语义上下文，增强理解基础
- 在WikiTQ上用GPT-4o-mini达78.13%准确率，优于现有方法
- 适合需要精准表格问答的开发者与研究者
表格是表示结构化关系数据的基本形式。尽管当前语言模型在文本任务中表现优异，但面对表格数据的复杂特性（如结构化特征）仍存在挑战。本文识别出四大关键问题：目标数据定位困难、表格语义缺失、文本推理中的数值错误以及符号推理的语义僵化。为此，我们提出TableMaster——一套综合框架与解决方案。该方法首先提取相关表格内容，并通过丰富语义上下文进行文字化表达；同时引入自适应推理机制，动态切换文本与符号推理，针对不同查询优化推理路径。大量分析与实验表明其有效性：在WikiTQ数据集上，使用GPT-4o-mini时达到78.13%准确率，显著超越现有基线。本工作为构建更鲁棒、可靠的表格理解系统提供了实用路径。
原文摘要 · Abstract (English)
Tables serve as a fundamental format for representing structured relational data. While current language models (LMs) excel at many text-based tasks, they still face challenges in table understanding due to the complex characteristics of tabular data, such as their structured nature. In this paper, we aim to enhance LMs for improved table understanding. We identify four key challenges: 1) difficulty in locating target data, 2) deficiency in table semantics, 3) numerical inaccuracies in textual reasoning, and 4) semantic inflexibility in symbolic reasoning. To address these issues, we propose TableMaster, a recipe and comprehensive framework that integrates multiple solutions to overcome these obstacles. TableMaster first extracts relevant table content and verbalizes it with enriched semantic context. Additionally, we introduce adaptive reasoning, a flexible approach that dynamically adjusts between textual and symbolic reasoning, tailoring the reasoning process to each query. Extensive analyses and experiments demonstrate our findings and the effectiveness of TableMaster. On the WikiTQ dataset, TableMaster achieves an accuracy of 78.13% using GPT-4o-mini, surpassing existing baselines. We hope this work will serve as a practical step toward more robust and reliable table understanding.
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。