# Emerging Vulnerabilities in Frontier Models: Multi-Turn Jailbreak Attacks

arXiv: https://arxiv.org/abs/2409.00137 · HTML: https://mlsanity.com/paper/2409.00137/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- Placed in Jailbreaking, one of 12 directions in LLM safety and security.
- Builds on 10 of its 15 references that are in the map.
- First posted on arXiv 29 Aug 2024.

## Placed in

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

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

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Jailbroken: How Does LLM Safety Training Fail?](https://mlsanity.com/paper/2307.02483/) (2023)
- [GPT-4 Is Too Smart To Be Safe: Stealthy Chat with LLMs via Cipher](https://mlsanity.com/paper/2308.06463/) (2023)
- [Great, Now Write an Article About That: The Crescendo Multi-Turn LLM Jailbreak Attack](https://mlsanity.com/paper/2404.01833/) (2024)
- [Exploiting Programmatic Behavior of LLMs: Dual-Use Through Standard Security Attacks](https://mlsanity.com/paper/2302.05733/) (2023)
- [LLM Defenses Are Not Robust to Multi-Turn Human Jailbreaks Yet](https://mlsanity.com/paper/2408.15221/) (2024)
- [Chain of Attack: a Semantic-Driven Contextual Multi-Turn attacker for LLM](https://mlsanity.com/paper/2405.05610/) (2024)
- [Speak Out of Turn: Safety Vulnerability of Large Language Models in Multi-turn Dialogue](https://mlsanity.com/paper/2402.17262/) (2024)
- [Token-Level Adversarial Prompt Detection Based on Perplexity Measures and Contextual Information](https://mlsanity.com/paper/2311.11509/) (2023)
- [When "Competency" in Reasoning Opens the Door to Vulnerability: Jailbreaking LLMs via Novel Complex Ciphers](https://mlsanity.com/paper/2402.10601/) (2024)

## Built on by (most cited first, 16 of 16)

- [AgentHarm: A Benchmark for Measuring Harmfulness of LLM Agents](https://mlsanity.com/paper/2410.09024/) (2024)
- [Accumulating Context Changes the Beliefs of Language Models](https://mlsanity.com/paper/2511.01805/) (2025)
- [Jigsaw Puzzles: Splitting Harmful Questions to Jailbreak Large Language Models](https://mlsanity.com/paper/2410.11459/) (2024)
- [You Know What I'm Saying: Jailbreak Attack via Implicit Reference](https://mlsanity.com/paper/2410.03857/) (2024)
- [Think Twice Before You Act: Enhancing Agent Behavioral Safety with Thought Correction](https://mlsanity.com/paper/2505.11063/) (2025)
- [Large Reasoning Models Are Autonomous Jailbreak Agents](https://mlsanity.com/paper/2508.04039/) (2025)
- [Prefill-level Jailbreak: A Black-Box Risk Analysis of Large Language Models](https://mlsanity.com/paper/2504.21038/) (2025)
- [Strategize Globally, Adapt Locally: A Multi-Turn Red Teaming Agent with Dual-Level Learning](https://mlsanity.com/paper/2504.01278/) (2025)
- [NEXUS: Network Exploration for eXploiting Unsafe Sequences in Multi-Turn LLM Jailbreaks](https://mlsanity.com/paper/2510.03417/) (2025)
- [Automating Deception: Scalable Multi-Turn LLM Jailbreaks](https://mlsanity.com/paper/2511.19517/) (2025)
- [GRAF: Multi-turn Jailbreaking via Global Refinement and Active Fabrication](https://mlsanity.com/paper/2506.17881/) (2025)
- [FERRET: Framework for Expansion Reliant Red Teaming](https://mlsanity.com/paper/2603.10010/) (2026)
- [The Salami Slicing Threat: Exploiting Cumulative Risks in LLM Systems](https://mlsanity.com/paper/2604.11309/) (2026)
- [Defensive M2S: Training Guardrail Models on Compressed Multi-turn Conversations](https://mlsanity.com/paper/2601.00454/) (2026)
- [Scalable Hierarchical Attention Transformers for Multi-Turn Jailbreak Detection in Long Conversations](https://mlsanity.com/paper/2606.21082/) (2026)
- [The Geometry of Harmfulness in Multi-Turn Attacks](https://mlsanity.com/paper/2609.38389/) (2026)

## Abstract (arXiv)

Large language models (LLMs) are improving at an exceptional rate. However, these models are still susceptible to jailbreak attacks, which are becoming increasingly dangerous as models become increasingly powerful. In this work, we introduce a dataset of jailbreaks where each example can be input in both a single or a multi-turn format. We show that while equivalent in content, they are not equivalent in jailbreak success: defending against one structure does not guarantee defense against the other. Similarly, LLM-based filter guardrails also perform differently depending on not just the input content but the input structure. Thus, vulnerabilities of frontier models should be studied in both single and multi-turn settings; this dataset provides a tool to do so.