Corpus ID: 220935513

Mapping natural-language problems to formal-language solutions using structured neural representations

  title={Mapping natural-language problems to formal-language solutions using structured neural representations},
  author={Kezhen Chen and Qiuyuan Huang and H. Palangi and P. Smolensky and Kenneth D. Forbus and Jianfeng Gao},
Generating formal-language programs represented by relational tuples, such as Lisp programs or mathematical operations, to solve problems stated in natural language is a challenging task because it requires explicitly capturing discrete symbolic structural information implicit in the input. However, most general neural sequence models do not explicitly capture such structural information, limiting their performance on these tasks. In this paper, we propose a new encoder-decoder model based on a… Expand
2 Citations


Question-Answering with Grammatically-Interpretable Representations
  • 35
  • PDF
Tensor Product Generation Networks for Deep NLP Modeling
  • 21
  • PDF
Attentive Tensor Product Learning
  • 9
  • PDF
Ain't Nobody Got Time For Coding: Structure-Aware Program Synthesis From Natural Language
  • 6
  • PDF
Learning From Unannotated QA Pairs to Analogically Disambiguate and Answer Questions
  • 8
  • PDF
MathQA: Towards Interpretable Math Word Problem Solving with Operation-Based Formalisms
  • 30
  • Highly Influential
  • PDF
Neural Program Search: Solving Programming Tasks from Description and Examples
  • 36
  • PDF
A Survey on Semantic Parsing
  • 32
  • PDF
A Compositional Neural Architecture for Language
  • A. E. Martin
  • Psychology, Medicine
  • Journal of Cognitive Neuroscience
  • 2020
  • 17
  • PDF
Tensor Product Variable Binding and the Representation of Symbolic Structures in Connectionist Systems
  • 409
  • PDF