Neural Code Completion
By Chang Liu, Xin Wang, Richard Shin, Joseph E. Gonzalez, Dawn Song. 2017
66 citations recorded by the source corpus.
Subfields: Natural Language Processing
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Natural Language Processing
- Charles Sutton
- Premkumar T. Devanbu
- Miltiadis Allamanis
- Ge Li
- Zhi Jin
- Earl T. Barr
- Tomasz Korbak
- Satish Chandra
Citations in the corpus
- A Survey of Machine Learning for Big Code and Naturalness (2017, background): “[185] and Yin and Neubig [196] use an LSTM over AST nodes to achieve the same goal.”
- CodeFill: Multi-token Code Completion by Jointly learning from Structure and Naming Sequences (2022, unknown): “Tomitigate rule-based autocompletion issues, researchers have proposed statistical [17, 37] and learning-based [6, 17, 29, 31, 32, 51] autocompletion models.”
- Energy-Based Models for Code Generation under Compilability Constraints (2021, background): “…constraints and use them for unconditional generation. tion (Ranzato et al., 2016) or actor critic (Konda and Tsitsiklis, 2000) for abstractive summarization (Paulus et al., 2018), caption generation (Liu et al., 2016b), dialogue (Li et al., 2016b), and video captioning (Pasunuru and Bansal, 2017).”
- Code Prediction by Feeding Trees to Transformers (2020, methodology): “The most common neural technique for code prediction is Recurrent Neural Networks (RNNs) [4] and their variants [5]–[10], where the code represented as a linear sequence is fed as input to the model.”
- Code Completion with Neural Attention and Pointer Networks (2017, methodology): “For the latter, we set the value vocabulary size to 50k to make the results comparable with (Liu et al. 2016), and employ all the three models.”
- Multi-task Learning based Pre-trained Language Model for Code Completion (2020, methodology): “ity, and dynamism in source code. In recent years, deep recurrent neural network-based language models have been applied to learning source code and have made great progress [3, 5, 27–29]. Liu et al. [28] proposed a code completion model based on a vanilla LSTM network. Li et al. [27] proposed a poi”