/CI-DD-Perses

MAPS(PLDI)'22: DD/Perses with CI

Primary LanguagePythonMIT LicenseMIT

CI-DD-Perses

This repository contains the artifacts of our paper 'Syntax-Guided Program Reduction for Understanding Neural Code Intelligence Models' accepted at MAPS'22 symposium, co-located with PLDI'22 conference.

Artifact for Article (CI-DD-Perses):

Reproducible Capsule of FeatureExtractor:


Structure

├── code/
    ├── CI_DD              # sample code for DD with CI
    ├── CI_Perses          # sample code for Perses with CI
    ├── files              # sample input program and model

├── data/
    ├── c2v_simplified     # traces of reduced inputs with code2vec model
    ├── c2s_simplified     # traces of reduced inputs with code2seq model
    ├── summary_results    # summary results of all experiments as csv

Approach

Extracting Key Features
Extracting key features that models have learned during training.

Citation:

Syntax-Guided Program Reduction for Understanding Neural Code Intelligence Models

@inproceedings{rabin2022features,
  author = {Rabin, Md Rafiqul Islam and Hussain, Aftab and Alipour, Mohammad Amin},
  title = {Syntax-Guided Program Reduction for Understanding Neural Code Intelligence Models},
  year = {2022},
  isbn = {9781450392730},
  publisher = {Association for Computing Machinery},
  address = {New York, NY, USA},
  url = {https://doi.org/10.1145/3520312.3534869},
  doi = {10.1145/3520312.3534869},
  booktitle = {Proceedings of the 6th ACM SIGPLAN International Symposium on Machine Programming},
  pages = {70–79},
  numpages = {10},
  location = {San Diego, CA, USA},
  series = {MAPS 2022}
}

Other Works: