Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat
Sign Up
Cited by 290

The DOI matches its Crossref record; 11 of the 11 references it cites were found in a registry or an index.

CURE: Code-Aware Neural Machine Translation for Automatic Program Repair

Open access
Jiang, Nan, Lutellier, Thibaud, Tan, Lin·2021·2021 IEEE/ACM 43rd International Conference on Software Engineering (ICSE)
DOI 10.1109/icse43902.2021.00107
Open this paper in CoChat — free account

Abstract

Automatic program repair (APR) is crucial to improve software reliability. Recently, neural machine translation (NMT) techniques have been used to automatically fix software bugs. While promising, these approaches have two major limitations. Their search space often does not contain the correct fix, and their search strategy ignores software knowledge such as strict code syntax. Due to these limitations, existing NMT-based techniques underperform the best template-based approaches. We propose CURE, a new NMT-based APR technique with three major novelties. First, CURE pre-trains a programming language (PL) model on a large software codebase to learn developer-like source code before the APR task. Second, CURE designs a new code-aware search strategy that finds more correct fixes by focusing on searching for compilable patches and patches that are close in length to the buggy code. Finally, CURE uses a subword tokenization technique to generate a smaller search space that contains more correct fixes. Our evaluation on two widely-used benchmarks shows that CURE correctly fixes 57 Defects4J bugs and 26 QuixBugs bugs, outperforming all existing APR techniques on both benchmarks.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/icse43902.2021.00107

Cites

11 of 11 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 11 references

Research works printed but not linked: 78

  1. murali, 2018, Neural Sketch Learning for Conditional Program Generation, ICLRE
  2. wang, 2017, Software reliability prediction using a deep learning model based on the RNN encoder–decoder, RESS
  3. 10.1109/QRS.2017.42 · DOI · Paper page
  4. 10.1145/3212695 · DOI · Paper page
  5. radford, 2019, Language Models are Unsupervised Multitask Learners, OpenAIRE blog, 1, 9
  6. 10.1109/MSR.2015.38 · DOI
  7. 10.1109/TSE.2020.3020502 · DOI
  8. 10.1145/3290353 · DOI · Paper page
  9. 10.1145/3293882.3330577 · DOI · Paper page
  10. 10.1613/jair.1.12007 · DOI · Paper page
  11. sutskever, 2014, Sequence to Sequence Learning with Neural Networks, NeurIPS, 3104
  12. radford, 2018, Improving Language Understanding by Generative Pre-Training, OpenAIRE blog
  13. vaswani, 2017, Attention Is All You Need, NeurIPS, 5998
  14. 10.1109/ICCIS.2013.490 · doi:10.1109/iccis.2013.490
  15. dinella, 2019, Hoppity: Learning Graph Transformations to Detect and Fix Bugs in Programs, ICLRE
  16. 10.1109/ICSME.2019.00050 · DOI
  17. tian, 2020, Evaluating Representation Learning of Code Changes for Predicting Patch Correctness in Program Repair, ASE
  18. 10.1145/3385412.3386005 · DOI
  19. 10.18653/v1/D19-5611 · DOI
  20. ghanbari, 2019, Practical Program Repair via Bytecode Mutation, ISSTA
  21. ding, 2020, Patching as Translation: the Data and the Metaphor, ASE
  22. 10.1145/3180155.3180167 · DOI · Paper page
  23. skorokhodov, 2018, Semi-Supervised Neural Machine Translation with Language Models, LoResMT AMTA, 37
  24. allamanis, 2016, A Convolutional Attention Network for Extreme Summarization of Source Code, ICML, 2091
  25. 10.1145/2970276.2970326 · DOI · Paper page
  26. 10.1109/ICSME.2017.46 · DOI
  27. 10.1145/2786805.2786811 · DOI · Paper page
  28. 10.1109/TSE.2011.104 · DOI · Paper page
  29. 10.1109/TSE.2019.2940179 · DOI · Paper page
  30. santos, 2017, Finding and Correcting Syntax Errors Using Recurrent Neural Networks, PeerJ PrePrints, 5, e3123v1
  31. 10.1145/3338906.3340455 · DOI
  32. gupta, 2017, Deepfix: Fixing Common C Language Errors by Deep Learning, AAAI, 1345
  33. 10.1145/2884781.2884872 · DOI · Paper page
  34. 10.1145/3238147.3240732 · DOI
  35. 10.1109/TSE.2018.2874648 · DOI
  36. 10.1007/978-3-319-99241-9_3 · DOI
  37. 10.1109/SANER.2019.8667991 · DOI
  38. 10.1145/3196398.3196473 · DOI · Paper page
  39. 10.1145/3377811.3380338 · DOI
  40. 10.1109/APSEC.2018.00085 · DOI
  41. 10.1007/s10664-019-09780-z · DOI · Paper page
  42. 10.1109/SANER.2019.8667970 · DOI · Paper page
  43. 10.1109/ICST.2019.00020 · DOI
  44. 10.1145/2896921.2896931 · DOI
  45. 10.1145/3236024.3264600 · DOI
  46. 10.1145/3106237.3106274 · DOI
  47. 10.1145/3180155.3180233 · DOI · Paper page
  48. 10.1109/SANER.2018.8330202 · DOI
  49. 10.1109/ASE.2017.8115674 · DOI
  50. 10.1109/SANER.2016.76 · DOI · Paper page
  51. 10.1109/TSE.2020.2979701 · DOI
  52. 10.1109/ICSE.2013.6606626 · DOI · Paper page
  53. henkel, 2018, Code Vectors: Under-standing Programs through Embedded Abstracted Symbolic Traces, ESEC/FSE, 163
  54. 10.1145/2884781.2884804 · DOI · Paper page
  55. 10.1007/978-3-030-36808-1_42 · DOI
  56. 10.1145/3377811.3380345 · DOI · Paper page
  57. 10.1145/3377811.3380361 · DOI · Paper page
  58. 10.1145/3395363.3397369 · DOI · Paper page
  59. alon, 2019, code2seq: Generating Sequences from Structured Representations of Code, ICLRE
  60. 10.1109/ASE.2019.00062 · DOI
  61. 10.1145/2568225.2568257 · DOI
  62. 10.1109/ASE.2017.8115675 · DOI · Paper page
  63. 10.1145/2931037.2948705 · DOI
  64. 10.1109/TSE.2016.2560811 · DOI · Paper page
  65. 10.1145/3385412.3385999 · DOI
  66. 10.1109/ASE.2017.8115676 · DOI
  67. wang, 2020, Learning Semantic Program Embeddings with Graph Interval Neural Network, ACM OOPSLA
  68. 10.1145/3106237.3106253 · DOI · Paper page
  69. 10.1007/978-3-030-58923-3_15 · DOI
  70. 10.1109/ASE.2019.00011 · DOI
  71. 10.1109/ICSE.2017.45 · DOI · Paper page
  72. 10.18653/v1/N19-4009 · DOI · Paper page
  73. wolf, 2020, Hugging-Face’s Transformers: State-of-the-art Natural Language Processing, EMNLP
  74. saha, 2019, Harnessing Evolution for MultiHunk Program Repair, ICSE IEEE, 13
  75. 10.1145/3135932.3135941 · DOI · Paper page
  76. 10.1109/MSR.2012.6224294 · DOI · Paper page
  77. 10.1109/ICSE.2015.139 · DOI · Paper page
  78. kingma, 2015, Adam: A Method for Stochastic Optimization, ICLRE

Cited by

See all 290

Metadata

Authors
Jiang, N., Lutellier, T., & Tan, L.
Venue
2021 IEEE/ACM 43rd International Conference on Software Engineering (ICSE)
Publisher
IEEE
Published
2021-05
Pages
1161-1173
DOI
10.1109/icse43902.2021.00107
OpenAlex
W3134686229
Cited by
290
References
11
FWCI
38.27
Type
conference paper
Language
en
Open access
green
Licence

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: CURE: Code-Aware Neural Machine Translation for Automatic Program Repair. https://app.cochat.ai/papers/10.1109/icse43902.2021.00107
The paper
Jiang, N., Lutellier, T., & Tan, L. (2021). CURE: Code-Aware Neural Machine Translation for Automatic Program Repair. 2021 IEEE/ACM 43rd International Conference on Software Engineering (ICSE). https://doi.org/10.1109/icse43902.2021.00107

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.