Automated program repair (APR) aims to help developers reduce the time and effort required for debugging and bug fixing by automatically generating patches for software defects. In recent years, APR has attracted growing interest from the software engineering, programming languages, and artificial intelligence communities. Despite major progress, important challenges remain, including improving patch quality, extending the applicability of repair techniques, increasing efficiency, and addressing usability and trustworthiness concerns for practitioners.
The International Workshop on Automated Program Repair (APR) provides a forum for researchers and practitioners to discuss new ideas, preliminary results, industrial experiences, and emerging directions in automated debugging and program repair. The workshop emphasizes both technical advances and human-centered aspects of repair, and seeks to strengthen connections between academia and industry to support broader adoption of APR technology.
Mon 12 OctDisplayed time zone: Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna change
09:00 - 10:00 | |||
09:00 60mKeynote | What Are We Really Repairing? Lessons from Cloud Failures APR Minghua Ma Microsoft | ||
12:30 - 14:00 | |||
14:00 - 16:00 | |||
14:00 20mResearch paper | Execution Features and Their Relations to Faults: A Replication on Defects4J APR Marius Smytzek CISPA Helmholtz Center for Information Security; Saarland University DOI | ||
14:20 20mShort-paper | Failure Explanations: The Missing Abstraction between Debugging and Repair APR Marius Smytzek CISPA Helmholtz Center for Information Security; Saarland University DOI | ||
14:40 20mResearch paper | AnchorRepair: A Reference-Guided Dual-Agent Framework for Code Translation Repair APR Jiajun Sun Tianjin University, Yaocai Zhao Tianjin University, Fengjie Li Tianjin University, Jiajun Jiang Tianjin University DOI | ||
15:00 20mResearch paper | RAWL: Retrieval-Augmented Smart Contract Repair using Weisfeiler-Lehman-Based Retrieval APR DOI | ||
Unscheduled Events
| Not scheduled Research paper | Evaluating Symbolic Feedback Granularity for LLM-Based Program Repair APR Taoer Guo New York University DOI |
Accepted Papers
Call for Papers
The International Workshop on Automated Program Repair (APR 2026) invites submissions on recent advances in the theory and practice of automated program repair, automated debugging, and closely related topics. The workshop will provide an opportunity for researchers and practitioners to exchange ideas, present emerging results, and discuss future research directions in the field.
Topics
Topics of interest include, but are not limited to:
- Quality of automatically generated patches
- Scalability of program repair algorithms
- Repair of domain-specific bugs
- Applications of program repair
- Deployment of program repair in practice
- Human aspects of program repair
- Case studies on program repair
- Program repair tools and benchmarks
- LLMs for program repair
- Dynamic and static analysis for program repair
- Fault localization, automated debugging, and program repair
- Testing and verification for program repair
- Program synthesis for program repair
- Test repair
- Program repair and DevOps
Submission Types
We welcome research papers (4-8 pages including references), position papers (2-4 pages including references), experience reports (4-8 pages including references) and talk proposals (1 page including references) from both academia and industry. The workshop aims to promote discussion on open challenges, practical deployment, and new directions for APR research.
Submission Format
All submissions must conform to the official ACM publication formats.
- Template: Submissions must use the official ACM Primary Article Template.
- LaTeX Users: Please use the sigconf format (i.e., \documentclass[sigconf,review,anonymous]{acmart}).
- Word Users: Please use the Word Interim Template.
- Submission should strictly follow the ACM Proceedings Template: https://www.acm.org/publications/proceedings-template
Double Anonymous Review Process
We will employ a double-anonymous review process. Thus, no submission may reveal its authors’ identities. The authors must make every effort to honor the double-anonymous review process. In particular:
- Authors’ names and affiliated institution names must be omitted from the submission.
- All references to the author’s prior work should be in the third person.
- Authors are encouraged to title their submissions differently than preprints of the authors on arXiV or similar sites. During review, authors should not publicly use the submission title.