Reasoning by Commented Code for Table Question Answering

TL;DR

This paper presents a step-by-step, commented code-generation approach for TableQA that improves reasoning clarity and accuracy, achieving state-of-the-art results on the WikiTableQuestions benchmark.

Contribution

It introduces a commented, multi-line code-generation framework with explicit reasoning comments, enhancing interpretability and accuracy in TableQA tasks.

Findings

Achieves 70.9% accuracy with Qwen2.5-Coder-7B-Instruct on WikiTableQuestions

Combining with end-to-end models boosts accuracy to 84.3%

Outperforms existing baseline methods in TableQA accuracy

Abstract

Table Question Answering (TableQA) poses a significant challenge for large language models (LLMs) because conventional linearization of tables often disrupts the two-dimensional relationships intrinsic to structured data. Existing methods, which depend on end-to-end answer generation or single-line program queries, typically exhibit limited numerical accuracy and reduced interpretability. This work introduces a commented, step-by-step code-generation framework that incorporates explicit reasoning into the Python program-generation process. The approach decomposes TableQA reasoning into multi-line executable programs with concise natural language comments, thereby promoting clearer reasoning and increasing the likelihood of generating correct code. On the WikiTableQuestions benchmark, the proposed method achieves 70.9\% accuracy using Qwen2.5-Coder-7B-Instruct, surpassing the Repanda baseline (67.6\%). Integrating the proposed framework with a robust end-to-end TableQA model via a lightweight answer-selection mechanism yields further improvements. This combined approach achieves up to 84.3\% accuracy on the WikiTableQuestions benchmark.