Does a language model “understand” high school math? A survey of deep learning based word problem solvers

Publikation: Beitrag in FachzeitschriftÜbersichtsarbeitForschungPeer-Review

Autoren

  • Sowmya S. Sundaram
  • Sairam Gurajada
  • Deepak Padmanabhan
  • Savitha Sam Abraham
  • Marco Fisichella

Organisationseinheiten

Externe Organisationen

  • Queen's University Belfast
  • Orebro University
  • IBM
Forschungs-netzwerk anzeigen

Details

OriginalspracheEnglisch
Aufsatznummere1534
Seitenumfang27
FachzeitschriftWiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery
Jahrgang14
Ausgabenummer4
PublikationsstatusVeröffentlicht - 11 Juli 2024

Abstract

From the latter half of the last decade, there has been a growing interest in developing algorithms for automatically solving mathematical word problems (MWP). It is a challenging and unique task that demands blending surface level text pattern recognition with mathematical reasoning. In spite of extensive research, we still have a lot to explore for building robust representations of elementary math word problems and effective solutions for the general task. In this paper, we critically examine the various models that have been developed for solving word problems, their pros and cons and the challenges ahead. In the last 2 years, a lot of deep learning models have recorded competing results on benchmark datasets, making a critical and conceptual analysis of literature highly useful at this juncture. We take a step back and analyze why, in spite of this abundance in scholarly interest, the predominantly used experiment and dataset designs continue to be a stumbling block. From the vantage point of having analyzed the literature closely, we also endeavor to provide a road-map for future math word problem research. This article is categorized under: Technologies > Machine Learning Technologies > Artificial Intelligence Fundamental Concepts of Data and Knowledge > Knowledge Representation.

ASJC Scopus Sachgebiete

Zitieren

Does a language model “understand” high school math? A survey of deep learning based word problem solvers. / Sundaram, Sowmya S.; Gurajada, Sairam; Padmanabhan, Deepak et al.
in: Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, Jahrgang 14, Nr. 4, e1534, 11.07.2024.

Publikation: Beitrag in FachzeitschriftÜbersichtsarbeitForschungPeer-Review

Sundaram, SS, Gurajada, S, Padmanabhan, D, Abraham, SS & Fisichella, M 2024, 'Does a language model “understand” high school math? A survey of deep learning based word problem solvers', Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, Jg. 14, Nr. 4, e1534. https://doi.org/10.1002/widm.1534
Sundaram, S. S., Gurajada, S., Padmanabhan, D., Abraham, S. S., & Fisichella, M. (2024). Does a language model “understand” high school math? A survey of deep learning based word problem solvers. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 14(4), Artikel e1534. https://doi.org/10.1002/widm.1534
Sundaram SS, Gurajada S, Padmanabhan D, Abraham SS, Fisichella M. Does a language model “understand” high school math? A survey of deep learning based word problem solvers. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery. 2024 Jul 11;14(4):e1534. doi: 10.1002/widm.1534
Sundaram, Sowmya S. ; Gurajada, Sairam ; Padmanabhan, Deepak et al. / Does a language model “understand” high school math? A survey of deep learning based word problem solvers. in: Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery. 2024 ; Jahrgang 14, Nr. 4.
Download
@article{cf017daf25bd451a9539b601dad6fc92,
title = "Does a language model “understand” high school math? A survey of deep learning based word problem solvers",
abstract = "From the latter half of the last decade, there has been a growing interest in developing algorithms for automatically solving mathematical word problems (MWP). It is a challenging and unique task that demands blending surface level text pattern recognition with mathematical reasoning. In spite of extensive research, we still have a lot to explore for building robust representations of elementary math word problems and effective solutions for the general task. In this paper, we critically examine the various models that have been developed for solving word problems, their pros and cons and the challenges ahead. In the last 2 years, a lot of deep learning models have recorded competing results on benchmark datasets, making a critical and conceptual analysis of literature highly useful at this juncture. We take a step back and analyze why, in spite of this abundance in scholarly interest, the predominantly used experiment and dataset designs continue to be a stumbling block. From the vantage point of having analyzed the literature closely, we also endeavor to provide a road-map for future math word problem research. This article is categorized under: Technologies > Machine Learning Technologies > Artificial Intelligence Fundamental Concepts of Data and Knowledge > Knowledge Representation.",
keywords = "automated word problem, deep learning, natural language processing, solving",
author = "Sundaram, {Sowmya S.} and Sairam Gurajada and Deepak Padmanabhan and Abraham, {Savitha Sam} and Marco Fisichella",
note = "Publisher Copyright: {\textcopyright} 2024 The Authors. WIREs Data Mining and Knowledge Discovery published by Wiley Periodicals LLC.",
year = "2024",
month = jul,
day = "11",
doi = "10.1002/widm.1534",
language = "English",
volume = "14",
number = "4",

}

Download

TY - JOUR

T1 - Does a language model “understand” high school math? A survey of deep learning based word problem solvers

AU - Sundaram, Sowmya S.

AU - Gurajada, Sairam

AU - Padmanabhan, Deepak

AU - Abraham, Savitha Sam

AU - Fisichella, Marco

N1 - Publisher Copyright: © 2024 The Authors. WIREs Data Mining and Knowledge Discovery published by Wiley Periodicals LLC.

PY - 2024/7/11

Y1 - 2024/7/11

N2 - From the latter half of the last decade, there has been a growing interest in developing algorithms for automatically solving mathematical word problems (MWP). It is a challenging and unique task that demands blending surface level text pattern recognition with mathematical reasoning. In spite of extensive research, we still have a lot to explore for building robust representations of elementary math word problems and effective solutions for the general task. In this paper, we critically examine the various models that have been developed for solving word problems, their pros and cons and the challenges ahead. In the last 2 years, a lot of deep learning models have recorded competing results on benchmark datasets, making a critical and conceptual analysis of literature highly useful at this juncture. We take a step back and analyze why, in spite of this abundance in scholarly interest, the predominantly used experiment and dataset designs continue to be a stumbling block. From the vantage point of having analyzed the literature closely, we also endeavor to provide a road-map for future math word problem research. This article is categorized under: Technologies > Machine Learning Technologies > Artificial Intelligence Fundamental Concepts of Data and Knowledge > Knowledge Representation.

AB - From the latter half of the last decade, there has been a growing interest in developing algorithms for automatically solving mathematical word problems (MWP). It is a challenging and unique task that demands blending surface level text pattern recognition with mathematical reasoning. In spite of extensive research, we still have a lot to explore for building robust representations of elementary math word problems and effective solutions for the general task. In this paper, we critically examine the various models that have been developed for solving word problems, their pros and cons and the challenges ahead. In the last 2 years, a lot of deep learning models have recorded competing results on benchmark datasets, making a critical and conceptual analysis of literature highly useful at this juncture. We take a step back and analyze why, in spite of this abundance in scholarly interest, the predominantly used experiment and dataset designs continue to be a stumbling block. From the vantage point of having analyzed the literature closely, we also endeavor to provide a road-map for future math word problem research. This article is categorized under: Technologies > Machine Learning Technologies > Artificial Intelligence Fundamental Concepts of Data and Knowledge > Knowledge Representation.

KW - automated word problem

KW - deep learning

KW - natural language processing

KW - solving

UR - http://www.scopus.com/inward/record.url?scp=85189518550&partnerID=8YFLogxK

U2 - 10.1002/widm.1534

DO - 10.1002/widm.1534

M3 - Review article

AN - SCOPUS:85189518550

VL - 14

JO - Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery

JF - Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery

SN - 1942-4787

IS - 4

M1 - e1534

ER -

Von denselben Autoren