TABBIE: Pretrained Representations of Tabular Data

TL;DR

TABBIE introduces a simple, efficient pretraining method for tabular data that outperforms joint table-text models on pure table tasks and provides comprehensive embeddings of table components.

Contribution

It proposes a novel pretraining objective for tabular data that improves performance on table-only tasks and offers detailed embeddings of table substructures.

Findings

Achieves state-of-the-art on table prediction tasks.

Provides embeddings for cells, rows, and columns.

Requires less compute than existing models.

Abstract

Existing work on tabular representation learning jointly models tables and associated text using self-supervised objective functions derived from pretrained language models such as BERT. While this joint pretraining improves tasks involving paired tables and text (e.g., answering questions about tables), we show that it underperforms on tasks that operate over tables without any associated text (e.g., populating missing cells). We devise a simple pretraining objective (corrupt cell detection) that learns exclusively from tabular data and reaches the state-of-the-art on a suite of table based prediction tasks. Unlike competing approaches, our model (TABBIE) provides embeddings of all table substructures (cells, rows, and columns), and it also requires far less compute to train. A qualitative analysis of our model's learned cell, column, and row representations shows that it understands complex table semantics and numerical trends.