English

FORTAP: Using Formulas for Numerical-Reasoning-Aware Table Pretraining

Information Retrieval 2022-03-29 v2 Machine Learning

Abstract

Tables store rich numerical data, but numerical reasoning over tables is still a challenge. In this paper, we find that the spreadsheet formula, which performs calculations on numerical values in tables, is naturally a strong supervision of numerical reasoning. More importantly, large amounts of spreadsheets with expert-made formulae are available on the web and can be obtained easily. FORTAP is the first method for numerical-reasoning-aware table pretraining by leveraging large corpus of spreadsheet formulae. We design two formula pretraining tasks to explicitly guide FORTAP to learn numerical reference and calculation in semi-structured tables. FORTAP achieves state-of-the-art results on two representative downstream tasks, cell type classification and formula prediction, showing great potential of numerical-reasoning-aware pretraining.

Keywords

Cite

@article{arxiv.2109.07323,
  title  = {FORTAP: Using Formulas for Numerical-Reasoning-Aware Table Pretraining},
  author = {Zhoujun Cheng and Haoyu Dong and Ran Jia and Pengfei Wu and Shi Han and Fan Cheng and Dongmei Zhang},
  journal= {arXiv preprint arXiv:2109.07323},
  year   = {2022}
}

Comments

Accepted by ACL'22 main track

R2 v1 2026-06-24T05:59:24.250Z