mlsanityData to 7 Oct 2026

Graph-based, Self-Supervised Program Repair from Diagnostic Feedback

arXiv 2005.10636 ↗First posted 20 May 2020cs.SE cs.CL cs.LGcited by 93

Not on the map. Why?

Abstract (arXiv)

We consider the problem of learning to repair programs from diagnostic feedback (e.g., compiler error messages). Program repair is challenging for two reasons: First, it requires reasoning and tracking symbols across source code and diagnostic feedback. Second, labeled datasets available for program repair are relatively small. In this work, we propose novel solutions to these two challenges. First, we introduce a program-feedback graph, which connects symbols relevant to program repair in source code and diagnostic feedback, and then apply a graph neural network on top to model the reasoning process. Second, we present a self-supervised learning paradigm for program repair that leverages unlabeled programs available online to create a large amount of extra program repair examples, which we use to pre-train our models. We evaluate our proposed approach on two applications: correcting introductory programming assignments (DeepFix dataset) and correcting the outputs of program synthesis (SPoC dataset). Our final system, DrRepair, significantly outperforms prior work, achieving 68.2% full repair rate on DeepFix (+22.9% over the prior best), and 48.4% synthesis success rate on SPoC (+3.7% over the prior best).

Key facts. Cited by 93 papers in the map (data through 7 Oct 2026). First posted on arXiv 20 May 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 19; 2022: 22; 2023: 28; 2024: 10; 2025: 6; 2026: 8.

93 papers.

Sort:

By area:

  1. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,304
  2. 2021Program Synthesis with Large Language Modelscited by 3,509
  3. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,317
  4. 2022Scaling Instruction-Finetuned Language Modelscited by 2,997
  5. 2023Teaching Large Language Models to Self-Debugcited by 963
  6. 2021Measuring Coding Challenge Competence With APPScited by 879
  7. 2023CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquingcited by 751
  8. 2022InCoder: A Generative Model for Code Infilling and Synthesiscited by 492
  9. 2021QA-GNN: Reasoning with Language Models and Knowledge Graphs for Question Answeringcited by 386
  10. 2021Graph Self-Supervised Learning: A Surveycited by 289
  11. 2022Generating Sequences by Learning to Self-Correctcited by 271
  12. 2022UnifiedSKG: Unifying and Multi-Tasking Structured Knowledge Grounding with Text-to-Text Language Modelscited by 231
38 more
  1. 2024When Can LLMs Actually Correct Their Own Mistakes? A Critical Survey of Self-Correction of LLMscited by 230
  2. 2023Is Self-Repair a Silver Bullet for Code Generation?cited by 191
  3. 2021Self-supervised Learning on Graphs: Contrastive, Generative,or Predictivecited by 136
  4. 2023From Word Models to World Models: Translating from Natural Language to the Probabilistic Language of Thoughtcited by 118
  5. 2023#InsTag: Instruction Tagging for Analyzing Supervised Fine-tuning of Large Language Modelscited by 108
  6. 2023Large Language Models as Analogical Reasonerscited by 107
  7. 2023The CoT Collection: Improving Zero-shot and Few-shot Learning of Language Models via Chain-of-Thought Fine-Tuningcited by 106
  8. 2023Learning Performance-Improving Code Editscited by 103
  9. 2023Execution-based Code Generation using Deep Reinforcement Learningcited by 95
  10. 2023LLMs cannot find reasoning errors, but can correct them given the error locationcited by 85
  11. 2022Coder Reviewer Reranking for Code Generationcited by 83
  12. 2021Leveraging Automated Unit Tests for Unsupervised Code Translationcited by 82
  13. 2021Break-It-Fix-It: Unsupervised Learning for Program Repaircited by 71
  14. 2022Repair Is Nearly Generation: Multilingual Program Repair with LLMscited by 68
  15. 2021MoCL: Data-driven Molecular Fingerprint via Knowledge-aware Contrastive Learning from Molecular Graphcited by 65
  16. 2021Learning to Synthesize Programs as Interpretable and Generalizable Policiescited by 59
  17. 2021A Syntax-Guided Edit Decoder for Neural Program Repaircited by 57
  18. 2023Bridging the Gap: A Survey on Integrating (Human) Feedback for Natural Language Generationcited by 55
  19. 2021Molecular Contrastive Learning with Chemical Element Knowledge Graphcited by 50
  20. 2023RepairLLaMA: Efficient Representations and Fine-Tuned Adapters for Program Repaircited by 50
  21. 2022Compilable Neural Code Generation with Compiler Feedbackcited by 46
  22. 2023SelfEvolve: A Code Evolution Framework via Large Language Modelscited by 43
  23. 2021JointLK: Joint Reasoning with Language Models and Knowledge Graphs for Commonsense Question Answeringcited by 40
  24. 2025A Survey of Test-Time Compute: From Intuitive Inference to Deliberate Reasoningcited by 40
  25. 2021Neural Transfer Learning for Repairing Security Vulnerabilities in C Codecited by 36
  26. 2021Learning to Repair: Repairing model output errors after deployment using a dynamic memory of feedbackcited by 35
  27. 2023Inference-Time Policy Adapters (IPA): Tailoring Extreme-Scale LMs without Fine-tuningcited by 33
  28. 2023Large Language Models of Code Fail at Completing Code with Potential Bugscited by 28
  29. 2023Teaching Language Models to Self-Improve through Interactive Demonstrationscited by 26
  30. 2023Deep Learning for Code Intelligence: Survey, Benchmark and Toolkitcited by 22
  31. 2021Learning Structural Edits via Incremental Tree Transformationscited by 16
  32. 2023ITER: Iterative Neural Repair for Multi-Location Patchescited by 16
  33. 2023B-Coder: Value-Based Deep Reinforcement Learning for Program Synthesiscited by 16
  34. 2023FLAME: A small language model for spreadsheet formulascited by 15
  35. 2022Neurosymbolic Repair for Low-Code Formula Languagescited by 12
  36. 2026M2F: Automated Formalization of Mathematical Literature at Scalecited by 12
  37. 2022CS1QA: A Dataset for Assisting Code-based Question Answering in an Introductory Programming Coursecited by 11
  38. 2022Improving Automated Program Repair with Domain Adaptationcited by 10

50 most cited of 93 All 93 (JSON)