Abstract
The task of math word problem (MWP) generation, which generates an MWP given an equation and relevant topic words, has increasingly attracted researchers’ attention. In this work, we introduce a simple memory retrieval module to search related training MWPs, which are used to augment the generation. To retrieve more relevant training data, we also propose a disentangled memory retrieval module based on the simple memory retrieval module. To this end, we first disentangle the training MWPs into logical description and scenario description and then record them in respective memory modules. Later, we use the given equation and topic words as queries to retrieve relevant logical descriptions and scenario descriptions from the corresponding memory modules, respectively. The retrieved results are then used to complement the process of the MWP generation. Extensive experiments and ablation studies verify the superior performance of our method and the effectiveness of each proposed module. The code is available at https://github.com/mwp-g/MWPG-DMR.
- [1] . 2019. Skeleton-to-response: Dialogue generation guided by retrieval memory. In Proceedings of the Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL’19).Google ScholarCross Ref
- [2] . 2019. Retrieval-guided dialogue response generation via a matching-to-generation framework. In Proceedings of the Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP’19). Association for Computational Linguistics, 1866–1875.
DOI: Google ScholarCross Ref - [3] . 2021. Neural machine translation with monolingual translation memory. In Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL’21). 7307–7318.Google ScholarCross Ref
- [4] . 2017. Reading wikipedia to answer open-domain questions. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). Association for Computational Linguistics, 1870–1879.
DOI: Google ScholarCross Ref - [5] . 2018. Isolating sources of disentanglement in variational autoencoders. Adv. Neural Inf. Process. Syst. 31 (2018).Google Scholar
- [6] . 2015. Microsoft coco captions: Data collection and evaluation server. arXiv:1504.00325. Retrieved from https://arxiv.org/abs/1504.00325Google Scholar
- [7] . 2003. Automatic item generation via frame semantics: Natural language generation of math word problems.Google Scholar
- [8] . 2018. A framework for the quantitative evaluation of disentangled representations. In International Conference on Learning Representations.Google Scholar
- [9] . 2016. Incorporating copying mechanism in sequence-to-sequence learning. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). Association for Computational Linguistics, 1631–1640.
DOI: Google ScholarCross Ref - [10] . 2018. Generating sentences by editing prototypes. Trans. Assoc. Comput. Ling. 6 (2018), 437–450.
DOI: Google ScholarCross Ref - [11] . 2020. REALM: Retrieval-augmented language model pre-training. In Proceedings of the 37th International Conference on Machine Learning (ICML’20). JMLR.org, Article
368 , 10 pages.Google Scholar - [12] . 2018. A retrieve-and-edit framework for predicting structured outputs. In Proceedings of the Conference and Workshop on Neural Information Processing Systems (NeurIPS’18).Google Scholar
- [13] . 2016. How well do computers solve math word problems? large-scale dataset construction and evaluation. In Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL’16). 887–896.Google ScholarCross Ref
- [14] . 2021. Recall and learn: A memory-augmented solver for math word problems. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP’21). Association for Computational Linguistics, 786–796.
DOI: Google ScholarCross Ref - [15] . 2021. Disenqnet: Disentangled representation learning for educational questions. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining. 696–704.Google ScholarDigital Library
- [16] . 2012. Retrieval-based learning: Active retrieval promotes meaningful learning. Curr. Direct. Psychol. Sci. 21 (2012), 157–163.Google ScholarCross Ref
- [17] . 2008. The critical importance of retrieval for learning. Science 319 (2008), 966–968.Google ScholarCross Ref
- [18] . 2020. Dense passage retrieval for open-domain question answering. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP’20). Association for Computational Linguistics, Online, 6769–6781.
DOI: Google ScholarCross Ref - [19] . 2020. Generalization through memorization: Nearest neighbor language models. In International Conference on Learning Representations.Google Scholar
- [20] . 2016. MAWPS: A math word problem repository. In Proceedings of the Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Association for Computational Linguistics, 1152–1157.
DOI: Google ScholarCross Ref - [21] . 2021. MWPToolkit: An open-source framework for deep learning-based math word problem solvers. arXiv:2109.00799. Retrieved from https://arxiv.org/abs/2109.00799Google Scholar
- [22] . 2007. METEOR: An automatic metric for MT evaluation with high levels of correlation with human judgments. In Proceedings of the 2nd Workshop on Statistical Machine Translation. 228–231.Google ScholarCross Ref
- [23] . 2019. Latent retrieval for weakly supervised open domain question answering. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. Association for Computational Linguistics, 6086–6096.
DOI: Google ScholarCross Ref - [24] . 2020. Pre-training via paraphrasing. In Proceedings of the 34th International Conference on Neural Information Processing Systems (NIPS’20). Curran Associates Inc., Red Hook, NY.Google ScholarDigital Library
- [25] . 2020. Retrieval-augmented generation for knowledge-intensive NLP tasks. In Advances in Neural Information Processing Systems, , , , , and (Eds.), Vol. 33. Curran Associates, Inc., 9459–9474.Google Scholar
- [26] . 2020. Graph-to-tree neural networks for learning structured input-output translation with applications to semantic parsing and math word problem. Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP’20).Google Scholar
- [27] . 2022. Hierarchical prediction and adversarial learning for conditional response generation. IEEE Trans. Knowl. Data Eng. 34, 1 (2022), 314–327.
DOI: Google ScholarDigital Library - [28] . 2022. Seeking patterns, not just memorizing procedures: Contrastive learning for solving math word problems. In Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL’22).Google Scholar
- [29] . 2021. MWP-BERT: Numeracy-augmented pre-training for math word problem solving. In Proceedings of the Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL-HLT’21).Google Scholar
- [30] . 2018. Memory-efficient and accurate sampling for counting local triangles in graph streams: From simple to multigraphs. ACM Trans. Knowl. Discov. Data 12, 1, Article
4 (Jan. 2018), 28 pages.DOI: Google ScholarDigital Library - [31] . 2020. Birds have four legs?! numersense: Probing numerical commonsense knowledge of pre-trained language models. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP’20). Association for Computational Linguistics, 6862–6868.
DOI: Google ScholarCross Ref - [32] . 2004. Rouge: A package for automatic evaluation of summaries. In Text Summarization Branches Out. 74–81.Google Scholar
- [33] . 2021. DiMBERT: Learning vision-language grounded representations with disentangled multimodal-attention. ACM Transactions on Knowledge Discovery from Data 16, 1, Article
1 (jul 2021), 19 pages.DOI: Google ScholarDigital Library - [34] . 2019. Tree-structured decoding for solving math word problems. In Proceedings of the Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP’19). Association for Computational Linguistics, 2370–2379.
DOI: Google ScholarCross Ref - [35] . 2020. Mathematical word problem generation from commonsense knowledge graph and equations. arXiv:2010.06196. Retrieved from https://arxiv.org/abs/2010.06196Google Scholar
- [36] . 2021. Mathematical word problem generation from commonsense knowledge graph and equations. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP’21).Google ScholarCross Ref
- [37] . 2019. Challenging common assumptions in the unsupervised learning of disentangled representations. In International Conference on Machine Learning. PMLR, 4114–4124.Google Scholar
- [38] . 2020. Knowledge Enhanced Neural Fashion Trend Forecasting. Association for Computing Machinery, New York, NY, 82–90.Google Scholar
- [39] . 2019. Who, where, and what to wear? extracting fashion knowledge from social media. In Proceedings of the 27th ACM International Conference on Multimedia (MM’19). Association for Computing Machinery, New York, NY, 257–265.Google ScholarDigital Library
- [40] . 2018. ATR-Vis: Visual and interactive information retrieval for parliamentary discussions in twitter. ACM Trans. Knowl. Discov. Data 12, 1, Article
3 (Feb. 2018), 33 pages.DOI: Google ScholarDigital Library - [41] . 2011. Math word question generation for training the students with learning difficulties. In Proceedings of the International Conference; Workshop on Emerging Trends in Technology. 206–211.Google ScholarDigital Library
- [42] . 2002. Bleu: A method for automatic evaluation of machine translation. In Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL’02). 311–318.Google Scholar
- [43] . 2014. Glove: Global vectors for word representation. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP’14). 1532–1543.Google ScholarCross Ref
- [44] . 2020. Disentangling by subspace diffusion. Adv. Neural Inf. Process. Syst. 33 (2020), 17403–17415.Google Scholar
- [45] . 2015. Personalized mathematical word problem generation. In Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI’15).Google Scholar
- [46] . 2023. Read, diagnose and chat: Towards explainable and interactive LLMs-augmented depression detection in social media. arXiv: 2305.05138. Retrieved from https://arxiv.org/abs/2305.05138Google Scholar
- [47] . 2019. Language models are unsupervised multitask learners. OpenAI blog 1 (2019), 9.Google Scholar
- [48] . 2010. Recent research on human learning challenges conventional instructional strategies. Educ. Res. 39 (2010), 406–412.Google ScholarCross Ref
- [49] . 2019. Multi-label punitive KNN with self-adjusting memory for drifting data streams. Proceedings of Machine Learning Research 13, 6, Article
60 (Nov. 2019), 31 pages.DOI: Google ScholarDigital Library - [50] . 2017. Get to the point: Summarization with pointer-generator networks. In Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL’17).Google ScholarCross Ref
- [51] . 2017. Attention is all you need. In Proceedings of the Conference and Workshop on Neural Information Processing Systems (NeurIPS’17).Google Scholar
- [52] . 2020. Word problems in mathematics education: A survey. ZDM 52 (2020), 1–16.Google ScholarCross Ref
- [53] . 2013. Using adaptive learning technologies to personalize instruction to student interests: The impact of relevant contexts on performance and learning outcomes. J. Educ. Psychol. 105 (2013), 932.Google ScholarCross Ref
- [54] . 2018. Mathdqn: Solving arithmetic word problems via deep reinforcement learning. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 32.Google ScholarCross Ref
- [55] . 2019. Template-based math word problem solvers with recursive neural networks. In Proceedings of the AAAI Conference on Artificial Intelligence, 7144–7151.
DOI: Google ScholarDigital Library - [56] . 2017. Knowledge graph embedding: A survey of approaches and applications. IEEE Trans. Knowl. Data Eng. 29, 12 (2017), 2724–2743.
DOI: Google ScholarCross Ref - [57] . 2017. Deep neural solver for math word problems. In Proceedings of the Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics, 845–854.
DOI: Google ScholarCross Ref - [58] . 2017. Deep neural solver for math word problems. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP’17). 845–854.Google ScholarCross Ref
- [59] . 2021. Math word problem generation with mathematical consistency and problem context constraints. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP’21).Google ScholarCross Ref
- [60] . 2018. Retrieve and refine: Improved sequence generation models for dialogue. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP’18).Google ScholarCross Ref
- [61] . 2011. Generating mathematical word problems. In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI’11).Google Scholar
- [62] . 2022. Graph neural networks for natural language processing: A survey. Found. Trends Mach. Learn. (2022).Google Scholar
- [63] . 2019. Response generation by context-aware prototype editing. In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI’19).Google ScholarDigital Library
- [64] . 2021. Knowledge graph embedding based on multi-view clustering framework. IEEE Trans. Knowl. Data Eng. 33, 2 (2021), 585–596.
DOI: Google ScholarDigital Library - [65] . 2019. A goal-driven tree-structured neural model for math word problems. In Proceedings of the 28th International Joint Conference on Artificial Intelligence (IJCAI’19). International Joint Conferences on Artificial Intelligence Organization, 5299–5305.
DOI: Google ScholarCross Ref - [66] . 2021. Deconfounded video moment retrieval with causal intervention. In Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. 1–10.Google ScholarDigital Library
- [67] . 2020. Teacher-student networks with multiple decoders for solving math word problem. In Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI’20).Google Scholar
- [68] . 2020. Do language embeddings capture scales?. In Findings of the Association for Computational Linguistics: Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP’20). Association for Computational Linguistics, 4889–4896.
DOI: Google ScholarCross Ref - [69] . 2019. Towards generating math word problems from equations and topics. In Proceedings of the 12th International Conference on Natural Language Generation (INLG’19).Google ScholarCross Ref
Index Terms
- Math Word Problem Generation via Disentangled Memory Retrieval
Recommendations
Math Word Problem Generation with Memory Retrieval
Pattern Recognition and Computer VisionAbstractThe task of math word problem generation (MWPG), which generates a math word problem (MWP) given an equation and several topic words, has increasingly attracted researchers’ attention. In this work, we propose a memory retrieval model to better ...
Information retrieval using word senses: root sense tagging approach
SIGIR '04: Proceedings of the 27th annual international ACM SIGIR conference on Research and development in information retrievalInformation retrieval using word senses is emerging as a good research challenge on semantic information retrieval. In this paper, we propose a new method using word senses in information retrieval: root sense tagging method. This method assigns coarse-...
Large Scale Retrieval and Generation of Image Descriptions
What is the story of an image? What is the relationship between pictures, language, and information we can extract using state of the art computational recognition systems? In an attempt to address both of these questions, we explore methods for ...
Comments