We gratefully acknowledge support from contributors and member institutions.
asixiv
···
Login
arXivarxiv:2607.11207Artificial Superintelligence

ProgramTab: Boosting Table Reasoning of LLMs via Programmatic Paradigm

Pei Guo, Enjie Liu, Yunzhi Tan, Mochi Gao, Jianxin Zhang, Ruichao Zhong, Juntao Li, Bo Hu, Zang Li

Table-based reasoning with large language models (LLMs), which requires reasoning based on natural language questions and structured tabular data, has gained widespread attention. However, a series of issues still constrain the application of this task. The previous approaches suffered from significant performance degradation when faced with large tables due to the difficulty of long text modeling and the limitation of input length for LLMs. The text-to-SQL approach is used to efficiently extract key information from tables and generate smaller sub-tables. However, tabular data, especially web tables, often lack the necessary structure and consistency, making them unsuitable for performing mathematical logic operations using SQL queries. We propose the ProgramTab framework, which guides LLMs employing in-context learning to perform tabular data preprocessing with Python code, as well as the momentous contents extraction with row and column extraction and SQL generation. The experiment results on table reasoning datasets demonstrate that the ProgramTab framework effectively deals with table-based reasoning tasks and outperforms all LLM-based baselines.

Subject:
asi
Submitted:
Jul 15, 2026
Views:
0