Incorporating External Knowledge through Pre-training for Natural Language to Code Generation

@article{Xu2020IncorporatingEK,
  title={Incorporating External Knowledge through Pre-training for Natural Language to Code Generation},
  author={F. F. Xu and Zhengbao Jiang and Pengcheng Yin and Bogdan Vasilescu and Graham Neubig},
  journal={ArXiv},
  year={2020},
  volume={abs/2004.09015}
}
Open-domain code generation aims to generate code in a general-purpose programming language (such as Python) from natural language (NL) intents. Motivated by the intuition that developers usually retrieve resources on the web when writing code, we explore the effectiveness of incorporating two varieties of external knowledge into NL-to-code generation: automatically mined NL-code pairs from the online programming QA forum StackOverflow and programming language API documentation. Our evaluations… Expand
7 Citations
In-IDE Code Generation from Natural Language: Promise and Challenges
  • PDF
Semantic Parsing with Less Prior and More Monolingual Data
  • Highly Influenced
  • PDF
Teach me how to Label: Labeling Functions from Natural Language with Text-to-text Transformers
  • PDF

References

SHOWING 1-10 OF 32 REFERENCES
Learning to Mine Aligned Code and Natural Language Pairs from Stack Overflow
  • 62
  • PDF
JuICe: A Large Scale Distantly Supervised Dataset for Open Domain Context-based Code Generation
  • 7
  • PDF
CoaCor: Code Annotation for Code Retrieval with Reinforcement Learning
  • 27
  • PDF
Language to Code: Learning Semantic Parsers for If-This-Then-That Recipes
  • 115
  • PDF
Latent Predictor Networks for Code Generation
  • 223
  • PDF
StaQC: A Systematically Mined Question-Code Dataset from Stack Overflow
  • 40
  • PDF
Neural Symbolic Machines: Learning Semantic Parsers on Freebase with Weak Supervision
  • 236
  • PDF
Mapping Language to Code in Programmatic Context
  • 46
  • PDF
Abstract Syntax Networks for Code Generation and Semantic Parsing
  • 196
  • PDF
...
1
2
3
4
...