CoChat
Sign Up
Cited by 24Software Engineering ResearchSoftware System Performance and ReliabilitySoftware Testing and Debugging Techniques

The DOI matches its Crossref record; 4 of the 4 references it cites were found in a registry or an index.

DeepDev-PERF: a deep learning-based approach for improving software performance

Garg, Spandan, Moghaddam, Roshanak Zilouchian, Clement, Colin B., Sundaresan, Neel, Wu, Chen·2022·Proceedings of the 30th ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering
DOI 10.1145/3540250.3549096
Open this paper in CoChat — free account

Abstract

Improving software performance is an important yet challenging part of the software development cycle. Today, the majority of performance inefficiencies are identified and patched by performance experts. Recent advancements in deep learning approaches and the wide-spread availability of open-source data creates a great opportunity to automate the identification and patching of performance problems. In this paper, we present DeepDev-PERF, a transformer-based approach to suggest performance improvements for C# applications. We pretrain DeepDev-PERF on English and Source code corpora, followed by finetuning for the task of generating performance improvement patches for C# applications. Our evaluation shows that our model can generate the same performance improvement suggestion as the developer fix in ‍53

Overview

Integrity signals

Registry

Crossref · 10.1145/3540250.3549096

Cites

4 of 4 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Verification

Claims the paper makes

This paper’s full text isn’t openly available, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Prior research has shown improvements in developer throughput with AI tools.

    “While some research has shown improvements in coding speed [12], developer throughput [13], and perceived productivity [14], more work must be done to validate these assertions across, for example, tasks, developer contexts, user groups, and more.”

    From How Much Does AI Impact Development Speed? an Enterprise-Based Randomized Controlled Trial (2025) · cites this paper as [13]

  • Current studies have missing nuance or understanding of the impact of AI tools by developer- or task-level contexts.

    “Other shortcomings include missing nuance or understanding of the impact of AI tools by developer- or task-level contexts (e.g., [13]), limited tool and task complexity in experimental settings (e.g., using GitHub Classroom rather than a more naturalistic developer environment [12]), and small sample sizes (e.g., only 32 participants in [15], 24 participants in [16], or 21 participants in [17]).”

    From How Much Does AI Impact Development Speed? an Enterprise-Based Randomized Controlled Trial (2025) · cites this paper as [13]

  • Cui et al.'s pooled analysis of three field experiments (n=4,867) found a 26% increase in throughput for developers using Copilot.

    “The authors found a 26% increase in throughput (measured as an increase in the number of pull requests) for developers using Copilot.”

    From How Much Does AI Impact Development Speed? an Enterprise-Based Randomized Controlled Trial (2025) · cites this paper as [13]

  • This estimate (21%) is significantly smaller than the 56% estimate shared by Peng and colleagues about GitHub Copilot [12], but aligns with the 26% productivity increase attributed to Copilot by Cui and colleagues [13].

    “This estimate is significantly smaller than the 56% estimate shared by Peng and colleagues about GitHub Copilot [12], but aligns with the 26% productivity increase attributed to Copilot by Cui and colleagues [13].”

    From How Much Does AI Impact Development Speed? an Enterprise-Based Randomized Controlled Trial (2025) · cites this paper as [13]

Citations

Cites

Flagged first, then by citations

Research works printed but not linked: 30

  1. 10.1007/978-3-319-63387-9_12 · DOI
  2. 10.1145/2568225.2568259 · DOI
  3. 10.18653/v1/2020.emnlp-main.728 · DOI
  4. 10.1145/2670979.2670987 · DOI
  5. 10.1145/2858965.2814290 · DOI
  6. Dawn Drain , Colin B. Clement , Guillermo Serrato , and Neel Sundaresan . 2021. DeepDebug: Fixing Python Bugs Using Stack Traces, Backtranslation, and Code Skeletons. ArXiv, abs/2105.09352 ( 2021 ), https://doi.org/10.48550/arXiv.2105.09352 10.48550/arXiv.2105.09352 Dawn Drain, Colin B. Clement, Guillermo Serrato, and Neel Sundaresan. 2021. DeepDebug: Fixing Python Bugs Using Stack Traces, Backtranslation, and Code Skeletons. ArXiv, abs/2105.09352 (2021), https://doi.org/10.48550/arXiv.2105.09352 · DOI
  7. 10.1145/1453101.1453111 · DOI
  8. 10.1609/aaai.v31i1.10742 · DOI · Paper page
  9. Md Shahriar Iqbal , Rahul Krishna , Mohammad Ali Javidian , Baishakhi Ray, and Pooyan Jamshidi. 2021 . CADET : Debugging and Fixing Misconfigurations using Counterfactual Reasoning . https://doi.org/10.48550/arXiv.2010.06061 10.48550/arXiv.2010.06061 Md Shahriar Iqbal, Rahul Krishna, Mohammad Ali Javidian, Baishakhi Ray, and Pooyan Jamshidi. 2021. CADET: Debugging and Fixing Misconfigurations using Counterfactual Reasoning. https://doi.org/10.48550/arXiv.2010.06061 · DOI
  10. 10.1145/2076021.2048081 · DOI
  11. 10.1109/TSE.2011.104 · DOI
  12. 10.1145/2048066.2048070 · DOI
  13. 10.1145/2837614.2837617 · DOI
  14. 10.1145/3360578 · DOI
  15. 10.1109/MSR.2013.6624035 · DOI
  16. Paolo Pialorsi and Marco Russo . 2007. Introducing microsoft® linq . Microsoft Press . Paolo Pialorsi and Marco Russo. 2007. Introducing microsoft® linq. Microsoft Press.
  17. Shuo Ren , Daya Guo , Shuai Lu , Long Zhou , Shujie Liu , Duyu Tang , M. Zhou , Ambrosio Blanco , and Shuai Ma. 2020. CodeBLEU: a Method for Automatic Evaluation of Code Synthesis. ArXiv, abs/2009.10297 ( 2020 ), https://doi.org/10.48550/arXiv.2009.10297 10.48550/arXiv.2009.10297 Shuo Ren, Daya Guo, Shuai Lu, Long Zhou, Shujie Liu, Duyu Tang, M. Zhou, Ambrosio Blanco, and Shuai Ma. 2020. CodeBLEU: a Method for Automatic Evaluation of Code Synthesis. ArXiv, abs/2009.10297 (2020), https://doi.org/10.48550/arXiv.2009.10297 · DOI · Paper page
  18. 10.1109/ICMLA.2018.00120 · DOI
  19. 10.1145/2660193.2660234 · DOI
  20. 10.1109/ICSE.2017.41 · DOI
  21. 10.1145/773473.178260 · DOI
  22. Michele Tufano , Dawn Drain , Alexey Svyatkovskiy , Shao Kun Deng, and Neel Sundaresan . 2020 . Unit Test Case Generation with Transformers. ArXiv , abs/2009.05617 (2020), https://doi.org/10.48550/arXiv.2009.05617 10.48550/arXiv.2009.05617 Michele Tufano, Dawn Drain, Alexey Svyatkovskiy, Shao Kun Deng, and Neel Sundaresan. 2020. Unit Test Case Generation with Transformers. ArXiv, abs/2009.05617 (2020), https://doi.org/10.48550/arXiv.2009.05617 · DOI
  23. 10.1145/3238147.3240732 · DOI
  24. 10.1109/ACSAC.2000.898880 · DOI
  25. 10.1145/2483760.2483784 · DOI
  26. 10.1145/1542476.1542523 · DOI
  27. 10.1145/1806596.1806617 · DOI
  28. 10.1145/1806596.1806616 · DOI
  29. 10.1007/978-3-642-16558-0_44 · DOI
  30. 10.1109/ICSE.2012.6227199 · DOI

Cited by

Sorted by citations

Metadata

Authors
Garg, S., Moghaddam, R. Z., Clement, C. B., Sundaresan, N., & Wu, C.
Affiliations
  • Microsoft (United States) (US)
  • Microsoft, China
ORCID
Venue
Proceedings of the 30th ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering
Publisher
ACM
Published
2022-11-07
Pages
948-958
DOI
10.1145/3540250.3549096
OpenAlex
W4308642031
Cited by
24
References
4
Citations by year
2026: 7 · 2025: 8 · 2024: 8 · 2023: 1
FWCI
3.80
Type
conference paper
Language
en
Open access
closed
Licence
www.acm.org/publications/policies/copyright_policy#Background (vor)
Locations
ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE) (publishedVersion)
Topics
  • Physical Sciences › Computer Science › Information Systems › Software Engineering Research
  • Physical Sciences › Computer Science › Computer Networks and Communications › Software System Performance and Reliability
  • Physical Sciences › Computer Science › Software › Software Testing and Debugging Techniques
Keywords
C# application

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