Reasoning arithmetic word problems entailing implicit relations based on the chain-of-thought model

Hao Meng; Lin Yue; Geng Sun; Jun Shen; Hao Meng; Lin Yue; Geng Sun; Jun Shen

doi:10.3934/steme.2023015

STEM Education

2023, Volume 3, Issue 4: 251-262. doi: 10.3934/steme.2023015

Previous Article Next Article

Research article Special Issues

Reasoning arithmetic word problems entailing implicit relations based on the chain-of-thought model

Hao Meng ^{1
,},
Lin Yue ^{2
,},
Geng Sun ^{3,4
,},
Jun Shen ^{5
,
,}

1.
School of Computing and Information Technology, University of Wollongong, Wollongong, Australia; hm578@uowmail.edu.au
2.
Department of Actuarial Studies and Business Analytics, Macquarie University, Sydney, Australia; lin.yue@mq.edu.au
3.
School of Computer Engineering, Chongqing College of Humanities, Science and Technology, Chongqing, China
4.
Vermilion Cloud, Sydney, Australia; gsun@vermilioncloud.com.au
5.
School of Computing and Information Technology, University of Wollongong, Wollongong, Australia; jshen@uow.edu.au

Academic Editor: Ergun Gide

Received: 21 August 2023 Revised: 19 October 2023 Accepted: 02 November 2023 Published: 15 November 2023

Solving arithmetic word problems ($ AWPs $) that involve deep implicit relations can be quite challenging. However, the paper proposed two approaches to tackle this issue. The first approach used the modifier-to-matrix ($ MTM $) model to extract noun modification components from the problem text. Specifically, a missing entity recovery ($ MER $) model translated explicit expressions into a node dependency graph ($ NDG $). The nodes on the graph then recursively acquired connections from the knowledge base through the $ MER $ model until the goal was achieved with a known quantity. The solving engine then selected the appropriate knowledge as the prompt. The second approach proposed in the paper was a comprehensive one that combined explicit and implicit knowledge to enhance reasoning abilities. The experimental results of the dataset demonstrate that the proposed algorithm is superior to the baseline algorithms in solving $ AWPs $ that require deep implicit relations.
- arithmetic word problem (AWP),
- implicit relation,
- chain-of-thought,
- missing entity recovery (MER),
- prompt
Citation: Hao Meng, Lin Yue, Geng Sun, Jun Shen. Reasoning arithmetic word problems entailing implicit relations based on the chain-of-thought model[J]. STEM Education, 2023, 3(4): 251-262. doi: 10.3934/steme.2023015

Related Papers:

Abstract

Solving arithmetic word problems ($ AWPs $) that involve deep implicit relations can be quite challenging. However, the paper proposed two approaches to tackle this issue. The first approach used the modifier-to-matrix ($ MTM $) model to extract noun modification components from the problem text. Specifically, a missing entity recovery ($ MER $) model translated explicit expressions into a node dependency graph ($ NDG $). The nodes on the graph then recursively acquired connections from the knowledge base through the $ MER $ model until the goal was achieved with a known quantity. The solving engine then selected the appropriate knowledge as the prompt. The second approach proposed in the paper was a comprehensive one that combined explicit and implicit knowledge to enhance reasoning abilities. The experimental results of the dataset demonstrate that the proposed algorithm is superior to the baseline algorithms in solving $ AWPs $ that require deep implicit relations.

References

[1]	Zhang, D., Wang, L., Zhang, L., Dai, B.T. and Shen, H.T., The gap of semantic parsing: A survey on automatic math word problem solvers. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2019, 42(9): 2287–2305. https://dx.doi.org/10.1109/TPAMI.2019.2914054 doi: 10.1109/TPAMI.2019.2914054
[2]	Sutskever, I., Vinyals, O. and Le, Q.V., Sequence to sequence learning with neural networks. Advances in neural information processing systems, 2014, 27. https://doi.org/10.48550/arXiv.1409.3215 doi: 10.48550/arXiv.1409.3215
[3]	Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., et al., Attention is all you need. Advances in neural information processing systems, 2017, 30.
[4]	Ling, W., Yogatama, D., Dyer, C. and Blunsom, P., Program induction by rationale generation Learning to solve and explain algebraic word problems. Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (ACL), 2017,158–167. https://dx.doi.org/10.18653/v1/P17-1015 doi: 10.18653/v1/P17-1015
[5]	Yang, K. and Deng, J., Learning to prove theorems via interacting with proof assistants. International Conference on Machine Learning (ICML), 2019, 97: 6984–6994.
[6]	Geva, M., Gupta, A. and Berant, J., Injecting numerical reasoning skills into language models. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (ACL), 2020,946–958.
[7]	Wei, J., Wang, X., Schuurmans, D., Bosma, M., Xia, F., Chi, E., et al., Chain of thought prompting elicits reasoning in large language models. Advances in Neural Information Processing Systems (NeurIPS), 2022, 35: 24824–24837.
[8]	Nakano, R., Hilton, J., Balaji, S., Wu, J., Ouyang, L., Kim, C., et al., Webgpt: Browser-assisted question-answering with human feedback. arXiv preprint arXiv: 2112.09332, 2021.
[9]	Zhou, D., Schärli, N., Hou, L., Wei, J., Scales, N., Wang, X., et al., Least-to-most prompting enables complex reasoning in large language models. International Conference on Learning Representations (ICLR), 2023.
[10]	Chen, W., Ma, X., Wang, X. and Cohen, W.W., Program of thought prompting: Disentangling computation from reasoning for numerical reasoning tasks. arXiv preprint arXiv: 2211.12588, 2022.
[11]	Wang, X., Wei, J., Schuurmans, D., Le, Q., Chi, E., Narang, S., et al., Self-consistency improves the chain of thought reasoning in language models. International Conference on Learning Representations (ICLR), 2023.
[12]	Li, Y., Lin, Z., Zhang, S., Fu, Q., Chen, B., Lou, J.G., et al., On the advance of making language models better reasoners. arXiv preprint arXiv: 2206.02336, 2022.
[13]	He, B., Meng, H., Zhang, Z., Liu, R., Zhang, T., Qualia Role-Based Quantity Relation Extraction for Solving Algebra Story Problems. CMES-Computer Modeling in Engineering & Sciences, 2023,136(1): 403–419. https://doi.org/10.32604/cmes.2023.023242 doi: 10.32604/cmes.2023.023242
[14]	Meng, H., Wu, H. and Yu, X., The Context-Oriented System Based on ELECTRA for Solving Math Word Problem. 2021 IEEE International Conference on Engineering, Technology & Education (TALE), 2021,976–981. https://dx.doi.org/10.1109/TALE52509.2021.9678762 doi: 10.1109/TALE52509.2021.9678762
[15]	Zhao, Y., Li, Y., Li, C. and Zhang, R., Multihiertt: Numerical reasoning over multi-hierarchical tabular and textual data. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (ACL), 2022, 6588–6600. https://dx.doi.org/10.18653/v1/2022.acl-long.454 doi: 10.18653/v1/2022.acl-long.454
[16]	Kojima, T., Gu, S.S., Reid, M., Matsuo, Y. and Iwasawa, Y., Large language models are zero-shot reasoners. Advances in neural information processing systems, 2022, 35: 22199–22213.
[17]	Khot, T., Trivedi, H., Finlayson, M., Fu, Y., Richardson, K., Clark, P., et al., Decomposed prompting: A modular approach for solving complex tasks. arXiv preprint arXiv: 2210.02406, 2022.
[18]	Gao, L., Madaan, A., Zhou, S., Alon, U., Liu, P., Yang, Y., et al., Pal: Program-aided language models. Proceedings of the 40th International Conference on Machine Learning, 2023,202: 10764–10799.
[19]	Lu, P., Peng, B., Cheng, H., Galley, M., Chang, K.W., Wu, Y.N., et al., Chameleon: Plug-and-play compositional reasoning with large language models. arXiv preprint arXiv: 2304.09842, 2023.

Author's biography Mr. Hao Meng is a joint Ph.D. candidate in Education Technology at the Central China Normal University (CCNU) and the University of Wollongong (UOW). His research interests include Intelligent Tutoring Systems, Technology Enhanced Learning, and Automated Problem Solver; Dr. Lin Yue is the Course Director of the Master of Business Analytics and Lecturer in the Department of Actuarial Studies and Business Analytics at Macquarie Business School. Her research is focused on intelligent networks, digital enablement, business / data analytics, strategic information systems and people analytics and modeling; Dr. Geng Sun is an adjunct professor at the School of Engineering, Chongqing College of Humanities, Science and Technology, China and the director of Vermilion Cloud, Australia. His current research interests and focus are education in the Web3 era, AI in education, intelligent tutoring systems, and computational intelligence in adaptive learning; Prof. Jun Shen is a Professor at the School of Computing and Information Technology, University of Wollongong, Wollongong, NSW, Australia. His expertise includes computational intelligence, bioinformatics, cloud computing, and learning technologies, including MOOC. He has published over 320 papers in journals and conferences in CS/IT areas

Reader Comments

Your name:*

Email:*
© 2023 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)