Robust Refusal Accuracy of Llama 3.1 and Mistral 7B Under Gradient-Based Adversarial Attacks

This report synthesises findings from 8 peer-reviewed papers addressing the following research question: What is the difference in robust refusal accuracy between Llama3.1 and Mistral 7B on the HarmBench dataset when subjected to gradient-based adversarial attacks. 16 claims were extracted from source literature; 11 were independently verified against retrieved documents. An automated multi-reviewer quality assessment produced a score of 7.5/10. This report is a machine-generated literature synthesis and does not constitute original research.

Research goal: What is the difference in robust refusal accuracy between Llama3.1 and Mistral 7B on the HarmBench dataset when subjected to gradient-based adversarial attacks?

Autonomous literature synthesis. Automated review score: 7.5/10. Full text and citation available at Assignee Research.