Filters








4,131 Hits in 5.2 sec

Enhancing the Transformer with Explicit Relational Encoding for Math Problem Solving [article]

Imanol Schlag, Paul Smolensky, Roland Fernandez, Nebojsa Jojic, Jürgen Schmidhuber, Jianfeng Gao
2020 arXiv   pre-print
We incorporate Tensor-Product Representations within the Transformer in order to better support the explicit representation of relation structure.  ...  Our Tensor-Product Transformer (TP-Transformer) sets a new state of the art on the recently-introduced Mathematics Dataset containing 56 categories of free-form math word-problems.  ...  Acknowledgments We thank the anonymous reviewers for their valuable comments. This research was supported by an European Research Council Advanced Grant (no: 742870).  ... 
arXiv:1910.06611v2 fatcat:z6ejzqytzrbgbdrk5yzxrztao4

Page 2955 of Mathematical Reviews Vol. , Issue 92e [page]

1992 Mathematical Reviews  
In order to solve these problems the authors consider channels with and without noiseless feedback. For each case they obtain an optimal encoding/decoding scheme and opti- mal transmission protocols.  ...  The zero-crossings of a wavelet transform define a representation which is well adapted for solving pattern recognition problems.  ... 

DISK: Domain-constrained Instance Sketch for Math Word Problem Generation [article]

Tianyang Cao, Shuang Zeng, Xiaodan Xu, Mairgup Mansur, Baobao Chang
2022 arXiv   pre-print
Besides, the QCG also interacts with the equation encoder to enhance the alignment between math tokens (e.g., quantities and variables) and MWP text.  ...  A math word problem (MWP) is a coherent narrative which reflects the underlying logic of math equations. Successful MWP generation can automate the writing of mathematics questions.  ...  This instance can be regarded as to constrain the scene with which the math problem text to be generated would be related.  ... 
arXiv:2204.04686v1 fatcat:6bm5igvwafcbtmjoelfzmzocqi

GeoQA: A Geometric Question Answering Benchmark Towards Multimodal Numerical Reasoning [article]

Jiaqi Chen, Jianheng Tang, Jinghui Qin, Xiaodan Liang, Lingbo Liu, Eric P. Xing, Liang Lin
2022 arXiv   pre-print
Automatic math problem solving has recently attracted increasing attention as a long-standing AI benchmark.  ...  Therefore, we propose a Geometric Question Answering dataset GeoQA, containing 4,998 geometric problems with corresponding annotated programs, which illustrate the solving process of the given problems  ...  Joint Reasoning Module Given the text feature H P and the diagram feature H D , it is crucial for solving geometric prob- H P is first encoded by 6 self-attention units (i.e., original Transformer), and  ... 
arXiv:2105.14517v3 fatcat:rhn2mugfwrbyfaz2ztk5d2dhy4

Page 3562 of Psychological Abstracts Vol. 90, Issue 11 [page]

2003 Psychological Abstracts  
The participants solved problems presented in a vertical format significantly faster than problems presented in a horizon- tal format.  ...  They also solved double digit first problems (e.g., 52+3) more quickly than the reverse (e.g., 3+52), but only when the problems were pre- sented horizontally.  ... 

MWP-BERT: Numeracy-Augmented Pre-training for Math Word Problem Solving [article]

Zhenwen Liang, Jipeng Zhang, Lei Wang, Wei Qin, Yunshi Lan, Jie Shao, Xiangliang Zhang
2022 arXiv   pre-print
Math word problem (MWP) solving faces a dilemma in number representation learning.  ...  In order to avoid the number representation issue and reduce the search space of feasible solutions, existing works striving for MWP solving usually replace real numbers with symbolic placeholders to focus  ...  To our knowledge, this is the first approach that surpasses human performance in terms of MWP solving. Related Works Math Word Problems Solving.  ... 
arXiv:2107.13435v2 fatcat:rbllb66w5fasllkpyp4nujz45i

The Gap of Semantic Parsing: A Survey on Automatic Math Word Problem Solvers [article]

Dongxiang Zhang, Lei Wang, Luming Zhang, Bing Tian Dai, Heng Tao Shen
2019 arXiv   pre-print
We also cover automatic solvers for other types of math problems such as geometric problems that require the understanding of diagrams.  ...  Finally, we identify several emerging research directions for the readers with interests in MWPs.  ...  Context-related Features The information embedded in the text window centered at a particular quantity can also provide important clues for solving math word problems.  ... 
arXiv:1808.07290v2 fatcat:spurptk2cjdw5lsqqejzfkfkqa

Stereotype-based stressors facilitate emotional memory neural network connectivity and encoding of negative information to degrade math self-perceptions among women

Chad E Forbes, Rachel Amey, Adam B Magerman, Kelly Duran, Mengting Liu
2018 Social Cognitive and Affective Neuroscience  
Findings provide an emotional memory encoding-based mechanism for well-established findings indicating that women have more negative math self-perceptions compared with men regardless of actual performance  ...  A total of 160 men and women completed a math test that provided veridical positive and negative feedback, a memory test for feedback, and math self-enhancing and valuing measures in a stereotype threatening  ...  Unexpectedly, stereotype threat did not uniquely enhance recall for information associated with solving problems incorrectly, at least behaviorally.  ... 
doi:10.1093/scan/nsy043 pmid:29939344 fatcat:fywmihthtfeq3e533cu5xyzzfe

Template-Based Math Word Problem Solvers with Recursive Neural Networks

Lei Wang, Dongxiang Zhang, Jipeng Zhang, Xing Xu, Lianli Gao, Bing Tian Dai, Heng Tao Shen
2019 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Then, we design a recursive neural network to encode the quantity with Bi-LSTM and self attention, and infer the unknown operator nodes in a bottom-up manner.  ...  The design of automatic solvers to arithmetic math word problems has attracted considerable attention in recent years and a large number of datasets and methods have been published.  ...  For example, (Wang, Liu, and Shi 2017) used LSTM as the encoder and GRU as the decoder for math problem solving.  ... 
doi:10.1609/aaai.v33i01.33017144 fatcat:abxav75phbbklljdx4hv4afqd4

Mining Numbers in Text: A Survey [chapter]

Minoru Yoshida, Kenji Kita
2021 Information Systems - Intelligent Information Processing Systems [Working Title]  
In this survey, we provide a quick overview of the history and recent advances of the research of mining such relations between numerals and words found in text data.  ...  However, relatively little attention has been paid in numerals found in texts and many systems treated the numbers found in the document in ad-hoc ways, such as regarded them as mere strings in the same  ...  Table 8 . 8 Numeracal Reading Comprehension Systems. 9 Table 9 . 9 Systems for Math Word Problem Solving. Data Table 10 . 10 Math Word Problem Datasets.  ... 
doi:10.5772/intechopen.98540 fatcat:mhzwc2ykcvcglfsh3ltxzvzq5q

Effects of Mathematical Word Problem—Solving Instruction on Middle School Students with Learning Problems

Yan Ping Xin, Asha K. Jitendra, Andria Deatline-Buchman
2005 The Journal of special education  
Next, students learned to transform the information in the diagram into a math sentence and solve for the unknown (Step 4). That is, they derived the following math equation, 12 ?  ...  type, and use the schema • Draw a picture to represent the problem diagram to represent the problemTransform the diagram to a math sentence, • Solve the problem and solve the problem • Look back to  ...  Sydney Zentall for her feedback on an earlier draft of this paper.  ... 
doi:10.1177/00224669050390030501 fatcat:pxzyryaw5vdgbdwltner3tlqzm

Translating Math Formula Images to LaTeX Sequences Using Deep Neural Networks with Sequence-level Training [article]

Zelun Wang, Jyh-Charn Liu
2019 arXiv   pre-print
To better capture the spatial relationships of math symbols, the feature maps are augmented with 2D positional encoding before being unfolded into a vector.  ...  The encoder is a convolutional neural network (CNN) that transforms images into a group of feature maps.  ...  The attention mechanism [23] has been introduced to solve this problem and has now become a widely adopted approach to enhance the performance on longer sequences.  ... 
arXiv:1908.11415v2 fatcat:3so7fujz4be7hluften6nbroqq

Making children gesture brings out implicit knowledge and leads to learning

Sara C. Broaders, Susan Wagner Cook, Zachary Mitchell, Susan Goldin-Meadow
2007 Journal of experimental psychology. General  
The authors found that, when told to gesture, children who were unable to solve the math problems often added new and correct problem-solving strategies, expressed only in gesture, to their repertoires  ...  The authors also found that when these children were given instruction on the math problems later, they were more likely to succeed on the problems than children told not to gesture.  ...  Noticing aspects of the world that are easily encoded in gesture seemed to help children learn how to solve the math problems in our study.  ... 
doi:10.1037/0096-3445.136.4.539 pmid:17999569 fatcat:hwzaavxgnbdije5zp45fuldfoy

Supporting Tutorial Feedback to Student Help Requests and Errors in Symbolic Differentiation [chapter]

Claus Zinn
2006 Lecture Notes in Computer Science  
SLOPERT's capabilities thus support the generation of natural problem-solving hints and scaffolding help.  ...  SLOPERT can provide natural step-by-step solutions for any given problem as well as diagnosis support for typical student errors.  ...  While these expert reasoning engines can solve a vast number of math problems, their algorithms are optimised for generality and efficiency, and thus, rarely mirror or mechanise human/student problem solving  ... 
doi:10.1007/11774303_35 fatcat:h4wlrwtctrhutafv35esilapiq

Mathematical Word Problem Generation from Commonsense Knowledge Graph and Equations [article]

Tianqiao Liu, Qiang Fang, Wenbiao Ding, Hang Li, Zhongqin Wu, Zitao Liu
2021 arXiv   pre-print
There is an increasing interest in the use of mathematical word problem (MWP) generation in educational assessment.  ...  The proposed model (1) learns both representations from edge-enhanced Levi graphs of symbolic equations and commonsense knowledge; (2) automatically fuses equation and commonsense knowledge information  ...  In order to well capture such relations, we model the edge la- bels as explicit nodes.  ... 
arXiv:2010.06196v3 fatcat:ej3adgz5wzcxzam24wl3xfnbza
« Previous Showing results 1 — 15 out of 4,131 results