Semantically-Aligned Universal Tree-Structured Solver for Math Word Problems

By Jinghui Qin, Lihui Lin, Xiaodan Liang, Rumin Zhang, Liang Lin. 2020

74 citations recorded by the source corpus.

A practical automatic textual math word problems (MWPs) solver should be able to solve various textual MWPs while most existing works only focused on one-unknown linear MWPs. Herein, we propose a simple but efficient method called Universal Expression Tree (UET) to make the first attempt to represent the equations of various MWPs uniformly. Then a semantically-aligned universal tree-structured solver (SAU-Solver) based on an encoder-decoder framework is proposed to resolve multiple types of MWPs in a unified model, benefiting from our UET representation. Our SAU-Solver generates a universal expression tree explicitly by deciding which symbol to generate according to the generated symbols' semantic meanings like human solving MWPs. Besides, our SAU-Solver also includes a novel subtree-level semanticallyaligned regularization to further enforce the semantic constraints and rationality of the generated expression tree by aligning with the contextual information. Finally, to validate the universality of our solver and extend the research boundary of MWPs, we introduce a new challenging Hybrid Math Word Problems dataset (HMWP), consisting of three types of MWPs. Experimental results on several MWPs datasets show that our model can solve universal types of MWPs and outperforms several state-of-the-art models.

Subfields: Artificial Intelligence, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Lei Wang
2. Xiaodan Liang
3. Pan Lu
4. Liang Lin
5. Nakul Verma
6. Bing Tian Dai
7. Ee-Peng Lim
8. Iddo Drori

In Natural Language Processing

1. Lei Wang
2. Iddo Drori
3. Nakul Verma
4. Bing Tian Dai
5. Pan Lu
6. Ee-Peng Lim
7. Yan Wang
8. Dongxiang Zhang

Citations in the corpus

• MWPToolkit: An Open-Source Framework for Deep Learning-Based Math Word Problem Solvers (2021, methodology): “…to solve MWPs, including Seq2Seq (Wang et al., 2017, 2018a; Chiang and Chen, 2019; Li et al., 2019), Seq2Tree (Wang et al., 2019; Liu et al., 2019a; Xie and Sun, 2019; Qin et al., 2020), Graph2Tree (Zhang et al., 2020b; Shen and Jin, 2020), and Pre-trained Language Models (Kim et al., 2020).”
• Are NLP Models really able to Solve Simple Math Word Problems? (2021, background): “HMWP (Qin et al., 2020) Poliak et al. (2018), and Guru-rangan et al. (2018).”
• Investigating Math Word Problems using Pretrained Multilingual Language Models (2021, methodology): “Similar approaches of evaluating MWPs difficulty levels are also used by Xie and Sun (2019) and Qin et al. (2020).”
• Neural-Symbolic Solver for Math Word Problems with Auxiliary Tasks (2021, methodology): “In this work, we deploy a tree-structured decoder (Xie and Sun, 2019) with attention mechanism (Bah-danau et al., 2015) as the backbone of our programmer and modify them with UET representation (Qin et al., 2020) to support more symbols for multiple types of MWPs.”
• A Survey of Deep Learning for Mathematical Reasoning (2022, unknown): “Additionally, there have been attempts to build non-English datasets for Chinese (Wang et al., 2017; Qin et al., 2020; Yu et al., 2021a) and Arabic (Alghamdi et al.”
• Solving Linear Algebra by Program Synthesis (2021, methodology): “Techniques have included sequence-tosequence and graph-to-tree Transformers which achieve around 80% on Math23k and MAWPS datasets (Koncel-Kedziorski et al., 2016; Li et al., 2019; Wang et al., 2019; Zhang et al., 2020; Li et al., 2020; Wu et al., 2020; Qin et al., 2020; Lan et al., 2021).”
• Inter-GPS: Interpretable Geometry Problem Solving with Formal Language and Symbolic Reasoning (2021, background): “The second line of work with a higher level of interpretability translates the math problems into symbolic language and conducts logical reasoning iteratively to predict the final results (Matsuzaki et al., 2017; Roy and Roth, 2018).”