Publications
Sort:
Open Access Issue
A graph-based fault localization method utilizing multi-modal code representations
Journal of Beijing University of Chemical Technology (Natural Science Edition) 2026, 53(5): 80-91
Published: 20 September 2026
Abstract PDF (904 KB) Collect
Downloads:0

Automatic fault localization techniques play a critical role in software development, but existing techniques lack in-depth characterization of program flow and code syntactic structure, making it difficult to understand deep logic and contextual information. They also lack universality in dealing with code entities of different granularity, limiting the information interaction of different levels of error localization and the comprehensive understanding of error characteristics.To addres these issues, this study proposes a graph-based fault localization framework utilizing multi-modal code representations, aiming to enhance fault localization effectiveness in two ways: firstly, leveraging multi-source information such as syntactic structure, semantic information, execution logic, and test coverage details to construct multi-modal code representation graphs using both large and small models, enhancing the ability to identify faults in complex code structures; secondly, forming multi-granularity localization models by sharing parameters between method and statement-level localization models to improve fault localization method accuracy. Experimental results conducted on 395 faults from the Defects4J dataset demonstrate that compared to the best baseline technique, our framework results in an increase of 25 correctly localized faults at the method level with a 1.12% improvement in mean first rank (MFR), and an increase of 34 correctly localized faults at the statement level with a 5.34% improvement in MFR.

Open Access Issue
A web test case repair approach based on test intent
Journal of Beijing University of Chemical Technology (Natural Science Edition) 2026, 53(4): 105-115
Published: 20 July 2026
Abstract PDF (6.4 MB) Collect
Downloads:2

Web test case automatic repair technology plays a crucial role in the maintenance of web application test suites. As web applications are frequently updated, changes may introduce web elements in new versions, causing test cases to fail. A major issue with existing methods is the lack of effective representation of the test intent and analysis of the reasons for failure, resulting in a failure to guide and assist in test case repair at a macro level. This makes it impossible to effectively repair broken test cases caused by test flow changes or propagated breakages. Empirical analysis shows that test intent and causes of failure are very useful for test case repair. This paper proposes a test intent-driven web test repair approach named LetTe (guiding LLM to fix Web UI tests based on test intent), which first parses the test intent and failure reasons of broken web test cases, and then guides the large language model (LLM) to fix them via prompt design and fine-tuning. LetTe’s repair logic simulates that of a human expert—given the test intent and reasons for failure, “ think about” possible repair plans and then generates repair candidates via the corresponding chain-of-thought. This paper evaluates LetTe across seven Web applications collected from open-source websites and publicly available datasets. The experimental results show that our approach achieves a 75% correct repair rate, which is higher than that of all baseline methods.

Total 2