# Safeguarding Efficacy in Large Language Models: Evaluating Resistance to Human-Written and Algorithmic Adversarial Prompts

arXiv: https://arxiv.org/abs/2510.15973 · HTML: https://mlsanity.com/paper/2510.15973/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Jailbreaking, one of 12 directions in LLM safety and security.
- Builds on 10 of its 23 references that are in the map.
- First posted on arXiv 12 Oct 2025.

## Placed in

- LLM safety and security / [Jailbreaking](https://mlsanity.com/safety/jailbreaking/)

## Builds on (most cited first, 10 of 10)

- [Universal and Transferable Adversarial Attacks on Aligned Language Models](https://mlsanity.com/paper/2307.15043/) (2023)
- [Jailbroken: How Does LLM Safety Training Fail?](https://mlsanity.com/paper/2307.02483/) (2023)
- [HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal](https://mlsanity.com/paper/2402.04249/) (2024)
- [RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models](https://mlsanity.com/paper/2009.11462/) (2020)
- [Ignore Previous Prompt: Attack Techniques For Language Models](https://mlsanity.com/paper/2211.09527/) (2022)
- [Tree of Attacks: Jailbreaking Black-Box LLMs Automatically](https://mlsanity.com/paper/2312.02119/) (2023)
- [Jailbreaking ChatGPT via Prompt Engineering: An Empirical Study](https://mlsanity.com/paper/2305.13860/) (2023)
- [Multi-step Jailbreaking Privacy Attacks on ChatGPT](https://mlsanity.com/paper/2304.05197/) (2023)
- [Survey of Vulnerabilities in Large Language Models Revealed by Adversarial Attacks](https://mlsanity.com/paper/2310.10844/) (2023)
- [Differential Robustness in Transformer Language Models: Empirical Evaluation Under Adversarial Text Attacks](https://mlsanity.com/paper/2509.09706/) (2025)

## Abstract (arXiv)

This paper presents a systematic security assessment of four prominent Large Language Models (LLMs) against diverse adversarial attack vectors. We evaluate Phi-2, Llama-2-7B-Chat, GPT-3.5-Turbo, and GPT-4 across four distinct attack categories: human-written prompts, AutoDAN, Greedy Coordinate Gradient (GCG), and Tree-of-Attacks-with-pruning (TAP). Our comprehensive evaluation employs 1,200 carefully stratified prompts from the SALAD-Bench dataset, spanning six harm categories. Results demonstrate significant variations in model robustness, with Llama-2 achieving the highest overall security (3.4% average attack success rate) while Phi-2 exhibits the greatest vulnerability (7.0% average attack success rate). We identify critical transferability patterns where GCG and TAP attacks, though ineffective against their target model (Llama-2), achieve substantially higher success rates when transferred to other models (up to 17% for GPT-4). Statistical analysis using Friedman tests reveals significant differences in vulnerability across harm categories (p < 0.001), with malicious use prompts showing the highest attack success rates (10.71% average). Our findings contribute to understanding cross-model security vulnerabilities and provide actionable insights for developing targeted defense mechanisms