Work

Reasoning and Retrieval for Complex Semi-structured Tables via Reinforced Relational Data Transformation

Year: 2025

Type: conference-paper

Abstract: We introduce TabFormer, a framework that normalizes diverse semi-structured tables into relational data via large language models to facilitate various table retrieval and reasoning tasks. Our approach employs a chain-of-thought methodology, transforming one or multiple tables through a sequence of soft operations. Compared to existing operators that are sensitive and brittle to human-induced artifacts in real-world tables, soft operators are designed with greater flexibility to accommodate diverse formatting variations.

Source: International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR)

DOI: 10.1145/3726302.3730071

Authors: Haoyu Dong, Yue Hu, Yanan Cao

Institutions: Chinese Academy of Sciences, Institute of Information Engineering, University of Chinese Academy of Sciences

Language: English

Topic: Advanced Database Systems and Queries

Subfield: Computer Networks and Communications

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

OpenAlex ID: W4412376860

FWCI: 2.599

Cites: 37

Cited by: 2

Related to: 10