.png){kind=icon}
Return
LibShift Search: A similarity-driven model for mapping deprecated Python API methods to current alternatives
.png){kind=icon}
PRE
AI
DOI：10.1016/j.infsof.2026.108305.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Source Code generation using Large Language Models (LLMs) has rapidly advanced, with newer models now capable of producing syntactically correct and functionally helpful code. However, a persistent issue remains: these models often suggest outdated or deprecated APIs due to their reliance on static, historical training data. As third-party libraries evolve rapidly, this mismatch poses challenges for real-world software development, particularly in maintaining code quality and compatibility with current best practices.
Journal
IF:
4.3
Papers:
3.8K
Citations:
7.7K
Organization
View more organizations
Cited Papers
No cited papers available