arXiv AI
Controlled Decoding Attacks on Black-Box LLMs
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv Computation and Language
Sep 23
Dynamic Deep Prompt Optimization for Defending Against Jailbreak Attacks on LLMs
arXiv:2609.26185v1 Announce Type: cross Abstract: Large Language Models (LLMs) demonstrate impressive capabilities across many applications but remain vulnerable to jailbreak attacks, which elicit ha...
arXiv AI
Jun 4
MaskForge: Structure-Aware Adaptive Attacks for Jailbreaking Diffusion Large Language Models
arXiv:2606. 04027v1 Announce Type: cross Abstract: Diffusion large language models (dLLMs) generate text by iteratively denoising partially masked sequences under bidirectional context, exposing a safety surface distinct from autoregressive LLMs.
arXiv Machine Learning
Jun 2
VERA: Variational Inference Framework for Jailbreaking Large Language Models
arXiv:2506. 22666v3 Announce Type: replace-cross Abstract: The rise of API-only access to state-of-the-art LLMs highlights the need for effective black-box jailbreak methods to identify model vulnerabilities in real-world settings.
arXiv Machine Learning
Jun 30
Sparse Autoencoders are Capable LLM Jailbreak Mitigators
arXiv:2602. 12418v2 Announce Type: replace-cross Abstract: Jailbreak attacks remain a persistent threat to large language model safety.
arXiv AI
Jun 19
SafeSpec: Fast and Safe LLM via Dynamic Reflective Sampling
arXiv:2606. 19755v1 Announce Type: cross Abstract: Speculative inference accelerates large language model (LLM) decoding but provides no inherent safety guarantees.
arXiv Computation and Language
Sep 21
CASCADE Against Jailbreaks: Combination Across Stages with Controlled Attack-Defense Evaluation
The paper presents a systematic study of combining defenses against jailbreak attacks on Large Language Models across different pipeline stages. It introduces a standardized evaluation framework that defines attack-success-rate, controls query budgets, and applies explicit fairness rules. Across 19 attacks and 15 defenses, the study finds that no single defense dominates, but carefully selected combinations can provide strong safety with minimal loss of utility, offering practical guidance for layered defense pipelines.