Adnan KhanAdnanProduct & AI
San Francisco Bay Area
Get in touch

Source note3 min read

Multi-Perspective Agentic Program Repair via Code Property Graphs and Temporal Execution Graphs

Source

arXiv

arxiv.org/abs/2607.12605v1

Read the source

Source

Plain-Language Brief

The paper introduces CT-Repair, an agentic Automated Program Repair (APR) framework that addresses the limitations of large language models in APR. The framework uses Code Property Graphs (CPG) and Temporal Execution Graphs (TEG) to represent static and dynamic evidence, and employs three finite-state-machine-guided agents for analysis from different perspectives. In an evaluation on 854 Java bugs, CT-Repair outperformed ReinFix and RepairAgent in repairing bugs under a controlled GPT-5.4-mini configuration, fixing 388 bugs compared to their respective 369 and 358. The framework also significantly reduces the amount of runtime evidence through filtering.

What the Document Says

  • CT-Repair is an automated program repair (APR) framework that uses Code Property Graphs (CPG) and Temporal Execution Graphs (TEG) to represent static and dynamic evidence.
  • The framework consists of three finite-state-machine-guided agents that analyze each bug from static, dynamic, and hybrid perspectives.
  • CT-Repair produces evidence-grounded repair strategies through a three-stage filtering pipeline.
  • The strategy-guided generation procedure instantiates these strategies as candidate patches and uses validation feedback to refine the most promising strategy.
  • In an evaluation on 854 Java bugs from Defects4J v3.0, CT-Repair correctly repairs 489 bugs in a mixed-model configuration, and 388 bugs under a controlled GPT-5.4-mini configuration.
  • CT-Repair outperforms ReinFix and RepairAgent, repairing 19 and 30 more bugs respectively.
  • The union of the three evidence perspectives repairs 99 more bugs than the strongest individual perspective.
  • The filtering pipeline compacts runtime evidence by narrowing the candidate method scope by 94.85% on average and reducing retained runtime records by 55.97%.

Key Concepts Mentioned

  • Automated Program Repair (APR)
  • Code Property Graph (CPG)
  • Temporal Execution Graph (TEG)
  • finite-state-machine
  • strategy-guided generation
  • filtering pipeline
  • Defects4J

Claims Made by the Document

  • CT-Repair improves repair effectiveness compared to other APR methods.
  • The three evidence perspectives in CT-Repair can provide more effective repair strategies than any individual perspective.

Evidence Used

  • Evaluation results on 854 Java bugs

Methods / Approach

  • Three-stage filtering pipeline
  • finite-state-machine-guided agents
  • strategy-guided generation
  • validation feedback

Limitations Stated by the Authors

Not stated in source

Limitations Noted by the Agent

Agent interpretation — not from the source.

The text does not provide information about potential limitations of CT-Repair.

Open Questions

  • What are the specific techniques used by each of the three finite-state-machine-guided agents?
  • How does CT-Repair handle bugs that are not covered in Defects4J v3.0?

Notes for Future Q&A

May include interpretation. Labeled where it does.

The paper does not delve into the specific techniques used by each agent or how it handles bugs outside of Defects4J.

Raw Abstract / Excerpt

Large language models (LLMs) have improved automated program repair (APR), but two limitations remain. First, raw execution traces are often too large and repetitive to serve as effective model context. Second, repeated patch sampling may produce different implementations without yielding distinct root-cause hypotheses or repair strategies. We present CT-Repair, an agentic APR framework representing static and dynamic evidence as queryable Code Property Graph (CPG) and Temporal Execution Graph (TEG). CT-Repair applies a three-stage filtering pipeline to construct compact TEGs. Three finite-state-machine-guided agents analyze each bug from static, dynamic, and hybrid perspectives and independently produce evidence-grounded repair strategies. A strategy-guided generation procedure instantiates these strategies as candidate patches and uses validation feedback to refine the most promising strategy. We evaluate CT-Repair on 854 Java bugs from Defects4J v3.0. In the mixed-model configuration, CT-Repair correctly repairs 489 bugs. Under a controlled GPT-5.4-mini configuration, it repairs 388 bugs, 19 and 30 more than ReinFix and RepairAgent, respectively. The union of the three evidence perspectives repairs 99 more bugs than the strongest individual perspective. The filtering pipeline also compacts runtime evidence, with execution filtering narrowing the candidate method scope by 94.85% on average and behavior filtering further reducing retained runtime records by 55.97%. These results show that structured runtime evidence and multi-perspective reasoning can improve repair effectiveness without relying solely on a larger patch-generation budget.