用逐行遍历提升表格推理准确率，无需训练且更高效。
RoT: Enhancing Table Reasoning with Iterative Row-Wise Traversals
- 通过逐行遍历表格并迭代反思，实现更可靠的推理过程。
- 不依赖训练，用非推理模型超越大语言模型平均4.3%。
- 减少幻觉、降低推理令牌数，适合资源受限场景使用。
表格推理任务对高效数据获取至关重要，旨在基于给定表格回答问题。近年来，具备长思维链（Long CoT）的大语言模型显著提升了推理能力，在表格推理上表现优异。然而，Long CoT存在训练成本高、表格内容幻觉严重等问题。为此，我们提出行思维（RoT），通过迭代式逐行遍历表格，实现推理扩展与反思式优化。通过行级遍历扩展推理长度，并利用大模型的反思能力，RoT实现零训练。顺序遍历增强对表格的关注，有效减少幻觉。实验表明，使用非推理模型的RoT平均优于推理大模型4.3%，在WikiTableQuestions和TableBench上达到当前最优水平，且相比Long CoT使用更少的推理令牌，展现出更高效率。
原文摘要 · Abstract (English)
The table reasoning task, crucial for efficient data acquisition, aims to answer questions based on the given table. Recently, reasoning large language models (RLLMs) with Long Chain-of-Thought (Long CoT) significantly enhance reasoning capabilities, leading to brilliant performance on table reasoning. However, Long CoT suffers from high cost for training and exhibits low reliability due to table content hallucinations. Therefore, we propose Row-of-Thought (RoT), which performs iteratively row-wise table traversal, allowing for reasoning extension and reflection-based refinement at each traversal. Scaling reasoning length by row-wise traversal and leveraging reflection capabilities of LLMs, RoT is training-free. The sequential traversal encourages greater attention to the table, thus reducing hallucinations. Experiments show that RoT, using non-reasoning models, outperforms RLLMs by an average of 4.3%, and achieves state-of-the-art results on WikiTableQuestions and TableBench with comparable models, proving its effectiveness. Also, RoT outperforms Long CoT with fewer reasoning tokens, indicating higher efficiency.
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。