@inproceedings{cheng2022fortap,
author = {Cheng, Zhoujun and Dong, Haoyu and Cheng, Fan and Jia, Ran and Wu, Pengfei and Han, Shi and Zhang, Dongmei},
title = {FORTAP: Using Formulae for Numerical-Reasoning-Aware Table Pretraining},
booktitle = {ACL 2022},
year = {2022},
month = {May},
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.},
url = {http://approjects.co.za/?big=en-us/research/publication/fortap-using-formulae-for-numerical-reasoning-aware-table-pretraining/},
}