Advances in Machine Learning for Vulnerability Detection and Antibiotic Resistance Prediction

The field of machine learning is moving towards ensemble learning and the use of large language models to improve the detection of vulnerabilities in source code and predict antibiotic resistance patterns. Researchers are exploring the potential of ensemble learning to create more robust vulnerability detection systems, and the results show that ensemble approaches can significantly improve detection performance. The use of large language models, such as CodeBERT and GraphCodeBERT, is also being investigated for vulnerability detection, and the results demonstrate their effectiveness. Additionally, the application of machine learning to predict antibiotic resistance patterns is a growing area of research, with studies showing promising results using techniques such as Sentence-BERT embeddings and XGBoost. Noteworthy papers in this area include: Ensembling Large Language Models for Code Vulnerability Detection, which proposes Dynamic Gated Stacking, a Stacking variant tailored for vulnerability detection. Predicting Antibiotic Resistance Patterns Using Sentence-BERT, which achieves an average F1 score of 0.86 using XGBoost.

Sources

"A 6 or a 9?": Ensemble Learning Through the Multiplicity of Performant Models and Explanations

Cross-Domain Evaluation of Transformer-Based Vulnerability Detection on Open & Industry Data

Ensembling Large Language Models for Code Vulnerability Detection: An Empirical Evaluation

Optimizing Code Embeddings and ML Classifiers for Python Source Code Vulnerability Detection

Predicting Antibiotic Resistance Patterns Using Sentence-BERT: A Machine Learning Approach

Built with on top of