Imagen de Portada

A Representation Engineering Perspective on the Effectiveness of Multi-Turn Jailbreaks

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Bullwinkel, Blake, Russinovich, Mark, Salem, Ahmed, Zanella-Beguelin, Santiago, Jones, Daniel, Severi, Giorgio, Kim, Eugenia, Hines, Keegan, Minnich, Amanda, Zunger, Yonatan, Kumar, Ram Shankar Siva
Formato:	Preprint
Publicado:	2025
Materias:	Cryptography and Security Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

The Trigger in the Haystack: Extracting and Reconstructing LLM Backdoor Triggers
por: Bullwinkel, Blake, et al.
Publicado: (2026)

GRP-Obliteration: Unaligning LLMs With a Single Unlabeled Prompt
por: Russinovich, Mark, et al.
Publicado: (2026)

A Systematization of Security Vulnerabilities in Computer Use Agents
por: Jones, Daniel, et al.
Publicado: (2025)

Great, Now Write an Article About That: The Crescendo Multi-Turn LLM Jailbreak Attack
por: Russinovich, Mark, et al.
Publicado: (2024)

Lessons From Red Teaming 100 Generative AI Products
por: Bullwinkel, Blake, et al.
Publicado: (2025)

Jailbreaking is (Mostly) Simpler Than You Think
por: Russinovich, Mark, et al.
Publicado: (2025)

Defending Against Indirect Prompt Injection Attacks With Spotlighting
por: Hines, Keegan, et al.
Publicado: (2024)

Securing AI Agents with Information-Flow Control
por: Costa, Manuel, et al.
Publicado: (2025)

Obliviate: Efficient Unmemorization for Protecting Intellectual Property in Large Language Models
por: Russinovich, Mark, et al.
Publicado: (2025)

Hey, That's My Model! Introducing Chain & Hash, An LLM Fingerprinting Technique
por: Russinovich, Mark, et al.
Publicado: (2024)

Optimizing Agent Planning for Security and Autonomy
por: Kolluri, Aashish, et al.
Publicado: (2026)

PyRIT: A Framework for Security Risk Identification and Red Teaming in Generative AI System
por: Munoz, Gary D. Lopez, et al.
Publicado: (2024)

Beyond Membership: Limitations of Add/Remove Adjacency in Differential Privacy
por: Pradhan, Gauri, et al.
Publicado: (2025)

LogiPlan: A Structured Benchmark for Logical Planning and Relational Reasoning in LLMs
por: Cai, Yanan, et al.
Publicado: (2025)

The Canary's Echo: Auditing Privacy Risks of LLM-Generated Synthetic Text
por: Meeus, Matthieu, et al.
Publicado: (2025)

Revisiting Jailbreaking for Large Language Models: A Representation Engineering Perspective
por: Li, Tianlong, et al.
Publicado: (2024)

MetaBackdoor: Exploiting Positional Encoding as a Backdoor Attack Surface in LLMs
por: Wen, Rui, et al.
Publicado: (2026)

Beyond the Single Turn: Reframing Refusals as Dynamic Experiences Embedded in the Context of Mental Health Support Interactions with LLMs
por: Tang, Ningjing, et al.
Publicado: (2026)

Closed-Form Bounds for DP-SGD against Record-level Inference
por: Cherubin, Giovanni, et al.
Publicado: (2024)

Many-Turn Jailbreaking
por: Yang, Xianjun, et al.
Publicado: (2025)

SEMA: Simple yet Effective Learning for Multi-Turn Jailbreak Attacks
por: Feng, Mingqian, et al.
Publicado: (2026)

TAMAÑO CORPORAL Y TEMPERATURA AMBIENTAL EN POBLACIONES CAZADORAS RECOLECTORAS DEL HOLOCENO TARDIO DE PAMPA Y PATAGONIA
por: Marien Béguelin
Publicado: (2010)

ESTIMACION DEL SEXO EN POBLACIONES DEL SUR DE SUDAMERCIA MEDIANTE FUNCIONES DISCRIMINANTES PARA EL FEMUR
por: Marien Béguelin
Publicado: (2008)

PONIENDO BLANCO SOBRE NEGRO: ANÁLISIS QUÍMICOS Y MICROSCÓPICOS SOBRE SEDIMENTOS Y RESTOS HUMANOS DE LA LAGUNA DEL JUNCAL (VALLE DEL RÍO NEGRO, NORPATAGONIA)
por: Marien Béguelin
Publicado: (2022)

ESTIMACION DE LA ESTATURA EN MUESTRAS DEL HOLOCENO TARDIO DEL N.O. DE SANTA CRUZ: PROBLEMAS METODOLOGICOS
por: Marien Béguelin
Publicado: (2005)

Estimación del sexo en cazadores-recolectores de Sudamérica a partir de variables métricas del húmero
por: Marien Béguelin
Publicado: (2011)

Variación morfométrica postcraneal en muestras tardías de restos humanos de Patagonia: una aproximación biogeográfica
por: Marien Béguelin
Publicado: (2006)

Permissive Information-Flow Analysis for Large Language Models
por: Siddiqui, Shoaib Ahmed, et al.
Publicado: (2024)

Not All Turns Matter: Credit Assignment for Multi-Turn Jailbreaking
por: He, Zhida, et al.
Publicado: (2026)

Preparative, Enactive, and Intertwined Theories of Change: Cultural Practitioners Influencing Conflict in Ecuador
por: Sarah Ullom-Minnich
Publicado: (2019)

Lightning detection rates and wildland fire in the mountains of northern Baja California, Mexico
por: Richard A. Minnich
Publicado: (1993)

The "Clipping Thesis": A Year Later.
por: Minnich, Nancy P.
Publicado: (1987)

Microforms & Secondary Schools
por: Minnich, Nancy P.
Publicado: (1972)

Computerized Literature Search Services in an Engineering Library
por: Hines, David L.
Publicado: (1975)

Multi-Turn Jailbreaks Are Simpler Than They Seem
por: Yang, Xiaoxue, et al.
Publicado: (2025)

The Echo Chamber Multi-Turn LLM Jailbreak
por: Alobaid, Ahmad, et al.
Publicado: (2026)

Automating Deception: Scalable Multi-Turn LLM Jailbreaks
por: Kumarappan, Adarsh, et al.
Publicado: (2025)

Projected Tensor-Tensor Products for Efficient Computation of Optimal Multiway Data Representations
por: Keegan, Katherine, et al.
Publicado: (2024)

Energy-lowering symmetry breaking creates a flat-band insulator in paramagnetic Nb3Cl8
por: Xiong, Jia-Xin, et al.
Publicado: (2024)

Chameleon: Increasing Label-Only Membership Leakage with Adaptive Poisoning
por: Chaudhari, Harsh, et al.
Publicado: (2023)