From LLMs to MLLMs to Agents: A Survey of Emerging Paradigms in Jailbreak Attacks and Defenses within LLM Ecosystem

Summary

This survey traces LLMs' development into multimodal LLMs and agents and examines the jailbreak attacks and defenses each stage introduces. It categorizes attacks by impact and attacker permissions, reviews representative methods, datasets and metrics, and organizes defenses by response timing and technical approach. It also identifies gaps in prior surveys, including agent-specific security, hybrid jailbreak taxonomy and experimental detail, and outlines future research directions.

Topics

• Prompt injection and jailbreaks

Related items

• MediumBeyond deterministic worms: Preparing the SOC for reasoning adversariesSimilar attack · CSO Online
• InfoShallowDefend: Harnessing Shallow Safety Alignment for Multi-Turn DefenderSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)
• InfoHow I Aligned Your Model: Scaling Up Multilingual Safety Training with TemplatesSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)
• InfoRed-Teaming Large Language Models in Low-Resource African Languages: A Case Study of Yoruba and Nigerian PidginSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)
• InfoSystematic Evaluation of Adversarial Jailbreaking Attacks and Defenses Across Seven Large Language ModelsSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)