mlsanityData to 7 Oct 2026

Detect-Localize-Repair: A Unified Framework for Learning to Debug with CodeT5

arXiv 2211.14875 ↗First posted 27 Nov 2022cs.SE cs.CLcited by 9

Not on the map. Why?

Abstract (arXiv)

Automated software debugging is a crucial task for improving the productivity of software developers. Many neural-based techniques have been proven effective for debugging-related tasks such as bug localization and program repair (or bug fixing). However, these techniques often focus only on either one of them or approach them in a stage-wise manner, ignoring the mutual benefits between them. In this work, we propose a novel unified Detect-Localize-Repair framework based on a pretrained programming language model CodeT5 to seamlessly address these tasks, named CodeT5-DLR. Specifically, we propose three objectives to adapt the generic CodeT5 for debugging: a bug detection objective to determine whether a given code snippet is buggy or not, a bug localization objective to identify the buggy lines, and a program repair objective to translate the buggy code to its fixed version. We evaluate it on each of these tasks and their combined setting on two newly collected line-level debugging datasets in Java and Python. Extensive results show that our model significantly outperforms existing baselines from both NLP and software engineering domains.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 26 of its 46 references that are in the map. First posted on arXiv 27 Nov 2022.

Builds on

26 of 46 references.

References · 26

  1. 2017Attention Is All You Needcited by 59,932
  2. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,266
  3. 2015Neural Machine Translation of Rare Words with Subword Unitscited by 2,898
  4. 2020CodeBERT: A Pre-Trained Model for Programming and Natural Languagescited by 1,286
  5. 2021CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generationcited by 890
  6. 2015Pointer Networkscited by 882
  7. 2021CodeXGLUE: A Machine Learning Benchmark Dataset for Code Understanding and Generationcited by 732
  8. 2020GraphCodeBERT: Pre-training Code Representations with Data Flowcited by 600
  9. 2021Unified Pre-training for Program Understanding and Generationcited by 361
  10. 2018SequenceR: Sequence-to-Sequence Learning for End-to-End Program Repaircited by 144
  11. 2020Graph-based, Self-Supervised Program Repair from Diagnostic Feedbackcited by 93
  12. 2021CURE: Code-Aware Neural Machine Translation for Automatic Program Repaircited by 85
14 more
  1. 2021Break-It-Fix-It: Unsupervised Learning for Program Repaircited by 71
  2. 2021Self-Supervised Bug Detection and Repaircited by 59
  3. 2021A Syntax-Guided Edit Decoder for Neural Program Repaircited by 57
  4. 2020Self-Supervised Contrastive Learning for Code Retrieval and Summarization via Semantic-Preserving Transformationscited by 49
  5. 2021Applying CodeBERT for Automated Program Repair of Java Simple Bugscited by 44
  6. 2019Neural Program Repair by Jointly Learning to Localize and Repaircited by 39
  7. 2015On the "Naturalness" of Buggy Codecited by 35
  8. 2019How Often Do Single-Statement Bugs Occur? The ManySStuBs4J Datasetcited by 35
  9. 2021How could Neural Networks understand Programs?cited by 26
  10. 2021Learning to Complete Code with Sketchescited by 23
  11. 2020Patching as Translation: the Data and the Metaphorcited by 18
  12. 2022On Distribution Shift in Learning-based Bug Detectorscited by 15
  13. 2022TSSB-3M: Mining single statement bugs at massive scalecited by 14
  14. 2018An Empirical Study of Fault Localization Families and Their Combinationscited by 13

Built on by

Cited by year

Citing papers per year: 2023: 2; 2024: 6; 2025: 1.

9 papers.

  1. 2024HyperAgent: Generalist Software Engineering Agents to Solve Coding Tasks at Scalecited by 61
  2. 2024When is Tree Search Useful for LLM Planning? It Depends on the Discriminatorcited by 46
  3. 2024AgileCoder: Dynamic Collaborative Agents for Software Development based on Agile Methodologycited by 45
  4. 2023RAP-Gen: Retrieval-Augmented Patch Generation with CodeT5 for Automatic Program Repaircited by 43
  5. 2023CodeTF: One-stop Transformer Library for State-of-the-art Code LLMscited by 21
  6. 2024CodeMMLU: A Multi-Task Benchmark for Assessing Code Understanding & Reasoning Capabilities of CodeLLMscited by 18
  7. 2024XMainframe: A Large Language Model for Mainframe Modernizationcited by 6
  8. 2025RepoDebug: Repository-Level Multi-Task and Multi-Language Debugging Evaluation of Large Language Modelscited by 4
  9. 2024S3LLM: Large-Scale Scientific Software Understanding with LLMs using Source, Metadata, and Documentcited by 2