Capturing Row and Column Semantics in Transformer Based Question Answering over Tables

TL;DR

This paper introduces two novel transformer-based models for table question answering that outperform existing methods without relying on specialized pre-training, achieving high accuracy and efficiency on benchmark datasets.

Contribution

The paper proposes two new approaches, RCI interaction and RCI representation, that improve accuracy and efficiency in table QA without specialized pre-training.

Findings

RCI interaction achieves ~98% Hit@1 accuracy on WikiSQL.

The interaction model outperforms TAPAS and TaBERT by 3.4% and 18.86%.

RCI representation offers significant efficiency advantages for online systems.

Abstract

Transformer based architectures are recently used for the task of answering questions over tables. In order to improve the accuracy on this task, specialized pre-training techniques have been developed and applied on millions of open-domain web tables. In this paper, we propose two novel approaches demonstrating that one can achieve superior performance on table QA task without even using any of these specialized pre-training techniques. The first model, called RCI interaction, leverages a transformer based architecture that independently classifies rows and columns to identify relevant cells. While this model yields extremely high accuracy at finding cell values on recent benchmarks, a second model we propose, called RCI representation, provides a significant efficiency advantage for online QA systems over tables by materializing embeddings for existing tables. Experiments on recent benchmarks prove that the proposed methods can effectively locate cell values on tables (up to ~98% Hit@1 accuracy on WikiSQL lookup questions). Also, the interaction model outperforms the state-of-the-art transformer based approaches, pre-trained on very large table corpora (TAPAS and TaBERT), achieving ~3.4% and ~18.86% additional precision improvement on the standard WikiSQL benchmark.