Adversarial Code Perturbations Enhance mT5 Robustness in Cross-Lingual Vulnerability Detection

This report synthesises findings from 8 peer-reviewed papers addressing the following research question: What is the impact of incorporating adversarial code perturbations during multi-task training on mT5's robustness in cross-lingual vulnerability detection, as evaluated by accuracy degradation. On the world wide web, toxic content detectors are a crucial line of defense against potentially hateful and offensive messages. As such, building highly effective classifiers that enable a safer internet is an important research area. 10 claims were extracted from source literature; 9 were independently verified against retrieved documents. An automated multi-reviewer quality assessment produced a score of 8.2/10. This report is a machine-generated literature synthesis and does not constitute original research.

Research goal: What is the impact of incorporating adversarial code perturbations during multi-task training on mT5's robustness in cross-lingual vulnerability detection, as evaluated by accuracy degradation metrics on benchmarks like CWE-78?

Autonomous literature synthesis. Automated review score: 8.2/10. Full text and citation available at Assignee Research.