.png){kind=icon}
.png){kind=icon}
Return
GraphClone: Structure-guided multi-view graph learning for code clone detection
.png){kind=icon}
PRE
AI
DOI：10.1016/j.jss.2026.113072.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Code clone detection is essential for various software engineering tasks, yet remains challenging due to the diverse implementations of functionally equivalent code. Existing studies have explored clone detection from different perspectives, including lexical/token-based matching, structure-aware analysis, and neural representation learning. Although these lines of research have substantially advanced the field, they still face limitations in capturing complementary syntactic, control-flow, and data-dependency information in a unified and efficient manner. These limitations may lead to suboptimal detection accuracy, especially for semantically similar but structurally diverse code fragments.
Journal
IF:
4.1
Papers:
5.4K
Citations:
8.4K
