ICSE 2026 (series) / Demonstrations /
PySTAAR: An End-to-End, Extensible Framework for Automated Python Type Error Repair
We present PySTAAR, an end-to-end, extensible framework for automatically detecting and repairing type errors in Python programs. PySTAAR integrates test generation, fault localization, patch synthesis, and patch validation into a fully automated pipeline, leveraging large language models and state-of-the-art repair techniques. The framework features a modular, extensible architecture and a user-friendly web interface that visualizes the workflow, making it accessible to developers of all experience levels. We demonstrate that PySTAAR can automatically detect and repair type errors in real-world Python applications, achieving high repair rate without manual intervention. Our tool demonstration is available at https://youtu.be/VizRQsrtsdk.
Fri 17 AprDisplayed time zone: Brasilia, Distrito Federal, Brazil change
Fri 17 Apr
Displayed time zone: Brasilia, Distrito Federal, Brazil change
11:00 - 12:30 | Testing and Analysis 17Demonstrations / Journal-first Papers / New Ideas and Emerging Results (NIER) at Oceania I Chair(s): Rangeet Pan IBM Research | ||
11:00 15mTalk | PySTAAR: An End-to-End, Extensible Framework for Automated Python Type Error Repair Demonstrations Wonseok Oh Korea University, Hyobin Park Kyungpook National University, Miryeong Kang Korea University, Seungbin Choi Kyungpook National University, Yunja Choi Kyungpook National University, Hakjoo Oh Korea University | ||
11:15 15mTalk | FlakeSync: A Tool for Automatically Repairing Async Flaky Tests Demonstrations Nandita Jayanthi The University of Texas at Austin, Shanto Rahman The University of Texas at Austin, August Shi The University of Texas at Austin | ||
11:30 15mTalk | The Sustainability Face of Automated Program Repair Tools Journal-first Papers Matias Martinez Universitat Politècnica de Catalunya (UPC), Silverio Martínez-Fernández UPC-BarcelonaTech, Xavier Franch Universitat Politècnica de Catalunya | ||
11:45 15mTalk | Towards Understanding the Challenges of Bug Localization in Deep Learning Systems Journal-first Papers Sigma Jahan Dalhousie University, Mehil Shah Dalhousie University, Masud Rahman Dalhousie University Pre-print File Attached | ||
12:00 15mTalk | Hypothesize-Then-Verify: Speculative Root Cause Analysis for Microservices with Pathwise Parallelism New Ideas and Emerging Results (NIER) Lingzhe Zhang Peking University, China, Tong Jia Institute for Artificial Intelligence, Peking University, Beijing, China, Yunpeng Zhai Alibaba Group, Leyi Pan Tsinghua University, Chiming Duan Peking University, Minghua He Peking University, Pei Xiao Peking University, Ying Li School of Software and Microelectronics, Peking University, Beijing, China | ||
12:15 15mTalk | Abductive Reasoning for Neurosymbolic Fault Localization New Ideas and Emerging Results (NIER) Minh Tam Le The University of Sydney, Australia, Xi Zheng Macquarie University, Hong Jin Kang University of Sydney | ||