mlsanityData to 7 Oct 2026

Rethinking Automated Program Repair: The Impact of Bug Complexity, Fault Localization, and LLM Cost-efficiency

arXiv 2608.14065 ↗First posted 14 Aug 2026cs.SE cs.AIcited by 1

Not on the map. Why?

Abstract (arXiv)

Background: Software bugs remain a critical challenge in development, necessitating effective Automated Program Repair (APR) techniques. While Large Language Model (LLM)-based APR systems have shown promise, prior studies primarily focus on overall repair effectiveness. The effects of bug complexity, fault localization, reasoning settings, and repair cost-effectiveness remain insufficiently explored. Aims: This study presents a comprehensive empirical analysis of LLM-based APR, focusing on how repair performance is shaped by bug complexity, fault localization, reasoning settings, and costs. Method: We evaluate two APR techniques (ChatRepair and CodeCorrector) using three LLMs (DeepSeek, GPT, and Llama), and examine their performance across diverse levels of bug complexity and localization strategies through a multi-dimensional empirical framework and statistical analysis. Results: Although structurally complex bugs and imprecise fault localization make repair more challenging, LLM-based APR techniques still achieve competitive repair effectiveness. Imprecise fault localization can substantially enlarge the performance gap between APR techniques. Furthermore, higher-cost LLMs and stronger reasoning settings do not consistently yield better cost-efficiency, revealing a nontrivial trade-off between repair effectiveness and computational cost. Conclusions: Over 50% of moderately complex bugs can be repaired by low-cost LLM-based APR techniques. The repair effectiveness gap between APR techniques becomes larger as fault localization becomes less precise. GPT-5 repairs 7 and 39 more complex bugs than DeepSeek-V4-pro and DeepSeek-V3.2, respectively; whereas the total repair cost of DeepSeek-V3.2 shows the best cost-efficiency performance.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 28 of its 51 references that are in the map. First posted on arXiv 14 Aug 2026.

Builds on

28 of 51 references.

References · 28

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  2. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  3. 2022Competition-Level Code Generation with AlphaCodecited by 1,427
  4. 2024A Survey on Large Language Models for Code Generationcited by 646
  5. 2023A Comprehensive Overview of Large Language Modelscited by 639
  6. 2024Token-Budget-Aware LLM Reasoningcited by 242
  7. 2024RepairAgent: An Autonomous, LLM-Based Agent for Program Repaircited by 225
  8. 2024The Effect of Sampling Temperature on Problem Solving in Large Language Modelscited by 168
  9. 2023Large Language Model Cascades with Mixture of Thoughts Representations for Cost-efficient Reasoningcited by 140
  10. 2023Keep the Conversation Going: Fixing 162 out of 337 bugs for 0.42 each using ChatGPTcited by 125
  11. 2023InferFix: End-to-End Program Repair with LLMscited by 113
  12. 2025Measuring what Matters: Construct Validity in Large Language Model Benchmarkscited by 102
16 more
  1. 2024A Survey on LLM-based Code Generation for Low-Resource and Domain-Specific Programming Languagescited by 65
  2. 2023Cost-Effective Hyperparameter Optimization for Large Language Model Generation Inferencecited by 40
  3. 2024A Deep Dive into Large Language Models for Automated Bug Localization and Repaircited by 39
  4. 2024Generative Artificial Intelligence: A Systematic Review and Applicationscited by 37
  5. 2023ChatGPT: Applications, Opportunities, and Threatscited by 28
  6. 2024Revisiting Code Similarity Evaluation with Abstract Syntax Tree Edit Distancecited by 26
  7. 2025Large Language Models for Code Generation: A Comprehensive Survey of Challenges, Techniques, Evaluation, and Applicationscited by 26
  8. 2024Training on the Benchmark Is Not All You Needcited by 24
  9. 2024ThinkRepair: Self-Directed Automated Program Repaircited by 22
  10. 2024RePair: Automated Program Repair with Process-based Feedbackcited by 16
  11. 2025Bag of Tricks for Inference-time Computation of LLM Reasoningcited by 15
  12. 2024Are Large Language Models Memorizing Bug Benchmarks?cited by 11
  13. 2018Automatic Repair of Real Bugs in Java: A Large-Scale Experiment on the Defects4J Datasetcited by 7
  14. 2024Automated Program Repair: Emerging trends pose and expose problems for benchmarkscited by 7
  15. 2024Using a Feedback Loop for LLM-based Infrastructure as Code Generationcited by 5
  16. 2025Evaluating the Generalizability of LLMs in Automated Program Repaircited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Does Fault Localization Beat a Fresh Attempt? A Placebo-Controlled Study of Test-Guided Code Repaircited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Large Language Models for Fault Localization: An Empirical Study ↗
  2. 2025The Art of Repair: Optimizing Iterative Program Repair with Instruction-Tuned Models
  3. 2026RGFL: Reasoning Guided Fault Localization for Automated Program Repair Using Large Language Models
  4. 2025Unlocking LLM Repair Capabilities Through Cross-Language Translation and Multi-Agent Refinement
  5. 2024An Empirical Study on LLM-based Agents for Automated Bug Fixing
  6. 2025A Survey of LLM-based Automated Program Repair: Taxonomies, Design Paradigms, and Applications
  7. 2025Automated Repair of C Programs Using Large Language Models
  8. 2025T3: Multi-level Tree-based Automatic Program Repair with Large Language Models