• Publications
  • Influence
Are deep neural networks the best choice for modeling source code?
TLDR
We present a fast, nested language modeling toolkit specifically designed for software, with the ability to add & remove text, and mix & swap out many models. Expand
  • 157
  • 24
  • PDF
Deep learning type inference
TLDR
We propose DeepTyper, a deep learning model that understands which types naturally occur in certain contexts and relations and can provide type suggestions, which can often be verified by the type checker, even if it could not infer the type initially. Expand
  • 68
  • 12
  • PDF
On the "naturalness" of buggy code
TLDR
We find that code with bugs tends to be more entropic (i.e. unnatural), becoming less so as bugs are fixed. Expand
  • 108
  • 8
Global Relational Models of Source Code
TLDR
We introduce two new hybrid model families that are both global and incorporate structural bias: Graph Sandwiches, which wrap traditional (gated) graph message-passing layers in sequential message- passing layers; and Graph Relational Embedding Attention Transformers, which bias traditional Transformers with relational information from graph edge types. Expand
  • 33
  • 8
Delft University of Technology On the “ Naturalness ” of Buggy Code
Real software, the kind working programmers produce by the kLOC to solve real-world problems, tends to be “natural”, like speech or natural language; it tends to be highly repetitive and predictable.Expand
  • 36
  • 5
Will They Like This? Evaluating Code Contributions with Language Models
TLDR
We find that rejected change sets do contain code significantly less similar to the project than accepted ones, furthermore, the less similar change sets are more likely to be subject to thorough review. Expand
  • 55
  • 3
  • PDF
CACHECA: A Cache Language Model Based Code Suggestion Tool
TLDR
CACHECA, an Eclipse plug in that combines the native suggestions with a statistical suggestion regime, more than doubles Eclipse's suggestion accuracy. Expand
  • 33
  • 2
  • PDF
On the naturalness of proofs
TLDR
We analyze proofs in two different proof assistant systems (Coq and HOL Light) to investigate if there is evidence of "naturalness" in these proofs. Expand
  • 5
  • 2
  • PDF
Revisiting Test Smells in Automatically Generated Tests: Limitations, Pitfalls, and Opportunities
TLDR
We benchmark the performance of two test smell detection tools: one widely used in prior work, and one recently introduced with the express goal to match developer perceptions of test smells. Expand
  • 2
  • 1
When Code Completion Fails: A Case Study on Real-World Completions
TLDR
This paper presents a case study on 15,000 code completions that were applied by 66 developers, which we study and contrast with artificial completions to inform future research and tools in this area. Expand
  • 13
  • PDF
...
1
2
...