@inproceedings{kim-etal-2020-point,
    title = "{P}oint to the {E}xpression: {S}olving {A}lgebraic {W}ord {P}roblems using the {E}xpression-{P}ointer {T}ransformer {M}odel",
    author = "Kim, Bugeun  and
      Ki, Kyung Seo  and
      Lee, Donggeon  and
      Gweon, Gahgene",
    editor = "Webber, Bonnie  and
      Cohn, Trevor  and
      He, Yulan  and
      Liu, Yang",
    booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
    month = nov,
    year = "2020",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/jlcl-multiple-ingestion/2020.emnlp-main.308/",
    doi = "10.18653/v1/2020.emnlp-main.308",
    pages = "3768--3779",
    abstract = "Solving algebraic word problems has recently emerged as an important natural language processing task. To solve algebraic word problems, recent studies suggested neural models that generate solution equations by using {\textquoteleft}Op (operator/operand)' tokens as a unit of input/output. However, such a neural model suffered two issues: expression fragmentation and operand-context separation. To address each of these two issues, we propose a pure neural model, Expression-Pointer Transformer (EPT), which uses (1) {\textquoteleft}Expression' token and (2) operand-context pointers when generating solution equations. The performance of the EPT model is tested on three datasets: ALG514, DRAW-1K, and MAWPS. Compared to the state-of-the-art (SoTA) models, the EPT model achieved a comparable performance accuracy in each of the three datasets; 81.3{\%} on ALG514, 59.5{\%} on DRAW-1K, and 84.5{\%} on MAWPS. The contribution of this paper is two-fold; (1) We propose a pure neural model, EPT, which can address the expression fragmentation and the operand-context separation. (2) The fully automatic EPT model, which does not use hand-crafted features, yields comparable performance to existing models using hand-crafted features, and achieves better performance than existing pure neural models by at most 40{\%}."
}