Paper deep dive

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

Zhili Huang, Ling Xu, Hongyu Zhang

Year: 2026Venue: arXiv preprintArea: cs.SEType: PreprintEmbeddings: 6

Intelligence

Status: succeeded | Model: Gemma-4-26B-A4B | Prompt: intel-v1 | Confidence: 93%

Last extracted: 7/18/2026, 11:09:17 AM

Summary

The paper introduces CT-Repair, an agentic Automated Program Repair (APR) framework that utilizes Code Property Graphs (CPG) and Temporal Execution Graphs (TEG) to represent static and dynamic evidence. It employs a three-stage filtering pipeline to compact runtime evidence and three finite-state-machine-guided agents to analyze bugs from static, dynamic, and hybrid perspectives. Evaluated on 854 Java bugs from Defects4J v3.0, CT-Repair outperforms ReinFix and RepairAgent, demonstrating that structured runtime evidence and multi-perspective reasoning improve repair effectiveness.

Entities (10)

CT-Repair · framework · 98%Code Property Graph · data-structure · 95%Hongyu Zhang · author · 95%Ling Xu · author · 95%Temporal Execution Graph · data-structure · 95%Zhili Huang · author · 95%Defects4J v3.0 · dataset · 92%ReinFix · baseline-model · 90%RepairAgent · baseline-model · 90%GPT-5.4 Mini · model · 85%

Relation Signals (8)

Ling Xu → authored → CT-Repair

confidence 95% · Authors:Zhili Huang, Ling Xu, Hongyu Zhang

Zhili Huang → authored → CT-Repair

confidence 95% · Authors:Zhili Huang ... View a PDF of the paper titled ... by Zhili Huang

Hongyu Zhang → authored → CT-Repair

confidence 95% · Authors:Zhili Huang, Ling Xu, Hongyu Zhang

CT-Repair → uses → Code Property Graph

confidence 95% · CT-Repair, an agentic APR framework representing static and dynamic evidence as queryable Code Property Graph (CPG)

CT-Repair → uses → Temporal Execution Graph

confidence 95% · CT-Repair, an agentic APR framework representing static and dynamic evidence as queryable ... Temporal Execution Graph (TEG)

CT-Repair → evaluatedon → Defects4J v3.0

confidence 92% · We evaluate CT-Repair on 854 Java bugs from Defects4J v3.0.

CT-Repair → outperforms → RepairAgent

confidence 90% · Under a controlled GPT-5.4-mini configuration, it repairs 388 bugs, ... 30 more than ... RepairAgent

CT-Repair → outperforms → ReinFix

confidence 90% · Under a controlled GPT-5.4-mini configuration, it repairs 388 bugs, 19 ... more than ReinFix

Cypher Suggestions (0)

No Cypher suggestions yet.

Abstract

Abstract: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.

PDF

Open source PDF →

PDF not stored locally. Use the link above to view on the source site.

Full Text

5,030 characters extracted from source content.

Expand or collapse full text

Skip to main content Search Submit Donate Log in Search arXiv Press Enter to search · Advanced search Computer Science > Software Engineering arXiv:2607.12605v1 (cs) [Submitted on 14 Jul 2026] Title:Multi-Perspective Agentic Program Repair via Code Property Graphs and Temporal Execution Graphs Authors:Zhili Huang, Ling Xu, Hongyu Zhang View a PDF of the paper titled Multi-Perspective Agentic Program Repair via Code Property Graphs and Temporal Execution Graphs, by Zhili Huang and 2 other authors View PDF Abstract: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. Comments: 12 pages, 5 figures, 10 tables Subjects: Software Engineering (cs.SE); Artificial Intelligence (cs.AI) ACM classes: D.2.5; D.2.7 Cite as: arXiv:2607.12605 [cs.SE] (or arXiv:2607.12605v1 [cs.SE] for this version) https://doi.org/10.48550/arXiv.2607.12605 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Zhili Huang [view email] [v1] Tue, 14 Jul 2026 10:33:29 UTC (6,015 KB) Full-text links: Access Paper: View a PDF of the paper titled Multi-Perspective Agentic Program Repair via Code Property Graphs and Temporal Execution Graphs, by Zhili Huang and 2 other authorsView PDFTeX Source view license Current browse context: cs.SE < prev | next > new | recent | 2026-07 Change to browse by: cs cs.AI References & Citations NASA ADSGoogle Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?) mathjaxToggle(); We gratefully acknowledge support from our major funders, member institutions, , and all contributors. About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab) Major funding support from

Instant research discovery