---
title: "Solving Math Word Problems with Teacher Supervision"
authors:
  - "Zhenwen Liang"
  - "Xiangliang Zhang"
doi: "10.24963/ijcai.2021/485"
url: "https://ijcai.org/proceedings/2021/485"
pdf: "https://ijcai.org/proceedings/2021/0485.pdf"
venue: "Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence"
year: 2021
pages: "3522-3528"
track: "Main Track"
keywords:
  - "Machine Learning Applications: Applications of Supervised Learning"
  - "Natural Language Processing: Question Answering"
---

# Solving Math Word Problems with Teacher Supervision

Zhenwen Liang, Xiangliang Zhang

*Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence*, pages 3522-3528. DOI: [10.24963/ijcai.2021/485](https://doi.org/10.24963/ijcai.2021/485)

## Abstract

Math word problems (MWPs) have been recently addressed with Seq2Seq models by `translating' math problems described in natural language to a mathematical expression, following a typical encoder-decoder structure. Although effective in solving classical math problems, these models fail when a subtle variation is applied to the word expression of a math problem, and leads to a remarkably different answer. We find the failure is because MWPs with different answers but similar math formula expression are encoded closely in the latent space. We thus designed a teacher module to make the MWP encoding vector match the correct solution and disaccord from the wrong solutions, which are manipulated from the correct solution. Experimental results on two benchmark MWPs datasets verified that our proposed solution outperforms the state-of-the-art models.