Image de couverture de livre

Mission Impossible: A Statistical Perspective on Jailbreaking LLMs

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Su, Jingtong, Kempe, Julia, Ullrich, Karen
Format:	Preprint
Publié:	2024
Sujets:	Machine Learning Artificial Intelligence Computation and Language
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers
par: Su, Jingtong, et autres
Publié: (2025)

Mission: Impossible Language Models
par: Kallini, Julie, et autres
Publié: (2024)

COLD-Attack: Jailbreaking LLMs with Stealthiness and Controllability
par: Guo, Xingang, et autres
Publié: (2024)

Efficient Safety Retrofitting Against Jailbreaking for LLMs
par: Garcia-Gasulla, Dario, et autres
Publié: (2025)

A Tale of Tails: Model Collapse as a Change of Scaling Laws
par: Dohmatob, Elvis, et autres
Publié: (2024)

A Single Character can Make or Break Your LLM Evals
par: Su, Jingtong, et autres
Publié: (2025)

Soft Tokens, Hard Truths
par: Butt, Natasha, et autres
Publié: (2025)

Embedding Trust: Semantic Isotropy Predicts Nonfactuality in Long-Form Text Generation
par: Bhardwaj, Dhrupad, et autres
Publié: (2025)

Understanding and Mitigating Tokenization Bias in Language Models
par: Phan, Buu, et autres
Publié: (2024)

JailbreakRadar: Comprehensive Assessment of Jailbreak Attacks Against LLMs
par: Chu, Junjie, et autres
Publié: (2024)

Iteration Head: A Mechanistic Study of Chain-of-Thought
par: Cabannes, Vivien, et autres
Publié: (2024)

The Impossibility Triangle of Long-Context Modeling
par: Zhou, Yan
Publié: (2026)

Jailbreaking LLMs via Calibration
par: Lu, Yuxuan, et autres
Publié: (2026)

Speak Easy: Eliciting Harmful Jailbreaks from LLMs with Simple Interactions
par: Chan, Yik Siu, et autres
Publié: (2025)

Best-of-N Jailbreaking
par: Hughes, John, et autres
Publié: (2024)

Jailbreaking with Universal Multi-Prompts
par: Hsu, Yu-Ling, et autres
Publié: (2025)

BiasJailbreak:Analyzing Ethical Biases and Jailbreak Vulnerabilities in Large Language Models
par: Lee, Isack, et autres
Publié: (2024)

Tree of Attacks: Jailbreaking Black-Box LLMs Automatically
par: Mehrotra, Anay, et autres
Publié: (2023)

Mission Impossible: Feedback-Guided Dynamic Interactive Planning for Improving Reasoning on LLMs
par: Yan, Dong, et autres
Publié: (2025)

Competition Report: Finding Universal Jailbreak Backdoors in Aligned LLMs
par: Rando, Javier, et autres
Publié: (2024)

A Decomposition Perspective to Long-context Reasoning for LLMs
par: Xiao, Yanling, et autres
Publié: (2026)

KDA: A Knowledge-Distilled Attacker for Generating Diverse Prompts to Jailbreak LLMs
par: Liang, Buyun, et autres
Publié: (2025)

Jailbreaking LLMs' Safeguard with Universal Magic Words for Text Embedding Models
par: Liang, Haoyu, et autres
Publié: (2025)

Are the Values of LLMs Structurally Aligned with Humans? A Causal Perspective
par: Kang, Yipeng, et autres
Publié: (2024)

Impossible Distillation: from Low-Quality Model to High-Quality Dataset & Model for Summarization and Paraphrasing
par: Jung, Jaehun, et autres
Publié: (2023)

Comprehensive Reassessment of Large-Scale Evaluation Outcomes in LLMs: A Multifaceted Statistical Approach
par: Sun, Kun, et autres
Publié: (2024)

Unveiling Downstream Performance Scaling of LLMs: A Clustering-Based Perspective
par: Xu, Chengyin, et autres
Publié: (2025)

Pruning for Protection: Increasing Jailbreak Resistance in Aligned LLMs Without Fine-Tuning
par: Hasan, Adib, et autres
Publié: (2024)

Shifting Perspectives: Steering Vectors for Robust Bias Mitigation in LLMs
par: Siddique, Zara, et autres
Publié: (2025)

LLMs can be Dangerous Reasoners: Analyzing-based Jailbreak Attack on Large Language Models
par: Lin, Shi, et autres
Publié: (2024)

How Data Inter-connectivity Shapes LLMs Unlearning: A Structural Unlearning Perspective
par: Qiu, Xinchi, et autres
Publié: (2024)

Instruction Learning Paradigms: A Dual Perspective on White-box and Black-box LLMs
par: Ren, Yanwei, et autres
Publié: (2025)

Leveraging the Context through Multi-Round Interactions for Jailbreaking Attacks
par: Cheng, Yixin, et autres
Publié: (2024)

Single-pass Detection of Jailbreaking Input in Large Language Models
par: Candogan, Leyla Naz, et autres
Publié: (2025)

Jailbreaking in the Haystack
par: Shah, Rishi Rajesh, et autres
Publié: (2025)

Configurable Foundation Models: Building LLMs from a Modular Perspective
par: Xiao, Chaojun, et autres
Publié: (2024)

Product of Experts with LLMs: Boosting Performance on ARC Is a Matter of Perspective
par: Franzen, Daniel, et autres
Publié: (2025)

DLF: Disentangled-Language-Focused Multimodal Sentiment Analysis
par: Wang, Pan, et autres
Publié: (2024)

Understanding Jailbreak Success: A Study of Latent Space Dynamics in Large Language Models
par: Ball, Sarah, et autres
Publié: (2024)

Principled RL for Diffusion LLMs Emerges from a Sequence-Level Perspective
par: Ou, Jingyang, et autres
Publié: (2025)