CRAILGOct 20, 2024

Jailbreaking and Mitigation of Vulnerabilities in Large Language Models

arXiv:2410.15236v239 citationsh-index: 10
Originality Synthesis-oriented
AI Analysis

It addresses security risks in LLMs for AI developers and users, but is incremental as it synthesizes existing research without new experimental results.

This review analyzes vulnerabilities in Large Language Models (LLMs), such as prompt injection and jailbreaking attacks, and presents defense strategies, categorizing attacks into prompt-based, model-based, multimodal, and multilingual types and evaluating defenses like prompt filtering and alignment techniques.

Large Language Models (LLMs) have transformed artificial intelligence by advancing natural language understanding and generation, enabling applications across fields beyond healthcare, software engineering, and conversational systems. Despite these advancements in the past few years, LLMs have shown considerable vulnerabilities, particularly to prompt injection and jailbreaking attacks. This review analyzes the state of research on these vulnerabilities and presents available defense strategies. We roughly categorize attack approaches into prompt-based, model-based, multimodal, and multilingual, covering techniques such as adversarial prompting, backdoor injections, and cross-modality exploits. We also review various defense mechanisms, including prompt filtering, transformation, alignment techniques, multi-agent defenses, and self-regulation, evaluating their strengths and shortcomings. We also discuss key metrics and benchmarks used to assess LLM safety and robustness, noting challenges like the quantification of attack success in interactive contexts and biases in existing datasets. Identifying current research gaps, we suggest future directions for resilient alignment strategies, advanced defenses against evolving attacks, automation of jailbreak detection, and consideration of ethical and societal impacts. This review emphasizes the need for continued research and cooperation within the AI community to enhance LLM security and ensure their safe deployment.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes