Instruction-Tuned Llama3 and DeepSeek-R1 Degradation Under Code Security Adversarial Perturbations

This report synthesises findings from 11 peer-reviewed papers addressing the following research question: Do instruction-tuned variants of Llama3 and Deepseek R1 exhibit different degradation patterns in helpfulness scores when subjected to taxonomy-specific adversarial perturbations in code security. 9 claims were extracted from source literature; 9 were independently verified against retrieved documents. An automated multi-reviewer quality assessment produced a score of 8.4/10. This report is a machine-generated literature synthesis and does not constitute original research.

Research goal: Do instruction-tuned variants of Llama3 and Deepseek R1 exhibit different degradation patterns in helpfulness scores when subjected to taxonomy-specific adversarial perturbations in code security tasks?

Autonomous literature synthesis. Automated review score: 8.4/10. Full text and citation available at Assignee Research.