WARM: A Weakly (+Semi) Supervised Model for Solving Math word Problems

TL;DR

This paper introduces WARM, a weakly and semi-supervised model for solving math word problems that only requires final answers for supervision, significantly reducing annotation effort and improving accuracy over previous methods.

Contribution

The paper proposes a novel weakly supervised approach for MWPs that learns to generate equations from problem descriptions and answers, and extends it to semi-supervised learning with new datasets.

Findings

Achieves 4.5% accuracy improvement on Math23K

Achieves 32% accuracy improvement on AllArith

Demonstrates effectiveness of weakly supervised learning

Abstract

Solving math word problems (MWPs) is an important and challenging problem in natural language processing. Existing approaches to solve MWPs require full supervision in the form of intermediate equations. However, labeling every MWP with its corresponding equations is a time-consuming and expensive task. In order to address this challenge of equation annotation, we propose a weakly supervised model for solving MWPs by requiring only the final answer as supervision. We approach this problem by first learning to generate the equation using the problem description and the final answer, which we subsequently use to train a supervised MWP solver. We propose and compare various weakly supervised techniques to learn to generate equations directly from the problem description and answer. Through extensive experiments, we demonstrate that without using equations for supervision, our approach achieves accuracy gains of 4.5% and 32% over the state-of-the-art weakly supervised approach, on the standard Math23K and AllArith datasets respectively. Additionally, we curate and release new datasets of roughly 10k MWPs each in English and in Hindi (a low resource language).These datasets are suitable for training weakly supervised models. We also present an extension of WARMM to semi-supervised learning and present further improvements on results, along with insights.