Shuo Liu

Vignat: Vulnerability identification by learning code semantics via graph attention networks

2023 · arXiv (Cornell University)

Vulnerability identification is crucial to protect software systems from attacks for cyber-security. However, huge projects have more than millions of lines of code, and the complex dependencies make it hard to carry out traditional static …

Delving into Parameter-Efficient Fine-Tuning in Code Change Learning: An Empirical Study

2024

Compared to Full-Model Fine-Tuning (FMFT), Parameter Efficient Fine-Tuning (PEFT) has demonstrated superior performance and lower computational overhead in several code understanding tasks, such as code summarization and code search. This advantage can be attributed to …

Exploration of Feature Engineering Teaching Based on Max-Relevance Min-Redundancy

2024

To meet the requirements of science and engineering personnel training in the new era, based on the teaching content of the feature engineering part of big data in the undergraduate teaching stage, combined with the …