AI Threat AlertLASH: Adaptive Semantic Hybridization for Black-Box Jailbreaking of Large Language Models
Jailbreak attacks expose a persistent gap between the intended safety behavior of aligned large language models and their behavior under adversarial prompting. Existing automated methods are increasingly effective but each
RedCodeAgent: Automatic Red-teaming Agent against Diverse Code Agents
they fail to cover certain boundary conditions, such as the combined effects of different jailbreak tools. In this work, we propose RedCodeAgent, the first automated red-teaming agent designed
PGuard: A Native, Streaming, and Plug-in Guardrail for Multimodal Large Language Models
based recurrent modules. Its design also accounts for the execution characteristics of LLM serving frameworks, such as SGLang, optimizing computation and data access to minimize additional runtime and synchronization overhead
PGuard: A Native, Streaming, and Plug-in Guardrail for Multimodal Large Language Models
based recurrent modules. Its design also accounts for the execution characteristics of LLM serving frameworks, such as SGLang, optimizing computation and data access to minimize additional runtime and synchronization overhead
An Evaluation of Data Leakage Risks in Tool-Using LLM Agents in Realistic Scenarios
leakage risks in agents has focused on adversarial data exfiltration through prompt injections and jailbreaks. However, sensitive information may also be exposed during non-adversarial use, creating leakage risks even
Claudini: Autoresearch Discovers State-of-the-Art Adversarial Attack Algorithms for LLMs
LLM agents like Claude Code can not only write code but also be used for autonomous AI research and engineering \citep{rank2026posttrainbench, novikov2025alphaevolve}. We show that an \emph{autoresearch}-style
A Fragile Guardrail: Diffusion LLM's Safety Blessing and Its Failure Mode
Diffusion large language models (D-LLMs) offer an alternative to
SAID: Empowering Large Language Models with Self-Activating Internal Defense
Large Language Models (LLMs), despite advances in safety alignment, remain vulnerable to jailbreak attacks designed to circumvent protective mechanisms. Prevailing defense strategies rely on external interventions, such as input filtering
TrapSuffix: Proactive Defense Against Adversarial Suffixes in Jailbreaking
inference pipeline. TrapSuffix channels jailbreak attempts into these two outcomes by reshaping the model's response landscape to adversarial suffixes. Across diverse suffix-based jailbreak settings, TrapSuffix reduces the average
Evaluating Adversarial Vulnerabilities in Modern Large Language Models
determined by the generation of disallowed content, with successful jailbreaks assigned a severity score. The findings indicate a disparity in jailbreak susceptibility between 2.5 Flash and GPT-4, suggesting variations
Vision-Language-Action Safety: Threats, Challenges, Evaluations, and Mechanisms
mitigated. We first define the scope of VLA safety, distinguishing it from text-only LLM safety and classical robotic safety, and review the foundations of VLA models, including architectures, training
GuardEval: A Multi-Perspective Benchmark for Evaluating Safety, Fairness, and Robustness in LLM Moderators
As large language models (LLMs) become deeply embedded in daily
@GrokSet: multi-party Human-LLM Interactions in Social Media
million tweets involving the @Grok LLM on X. Our analysis reveals a distinct functional shift: rather than serving as a general assistant, the LLM is frequently invoked as an authoritative
SafeCap: Improving LVLM Safety with Image Captioning Reinforcement Learning
Large vision-language models (LVLMs) remain vulnerable to jailbreak attacks that exploit visual inputs to bypass safety alignment inherited from their language backbones. We propose SafeCap, a reinforcement-learning framework
What Intermediate Layers Know: Detecting Jailbreaks from Entropy Dynamics
Jailbreak attacks reveal a persistent weakness in aligned Large Language Models: carefully crafted prompts can elicit policy-violating responses despite safety training. While most defenses operate at the prompt
Can LLMs Threaten Human Survival? Benchmarking Potential Existential Threats from LLMs via Prefix Completion
safety evaluation of large language models (LLMs) has become extensive, driven by jailbreak studies that elicit unsafe responses. Such response involves information already available to humans, such as the answer
Steering Vectors are an Adversarial Attack Surface
Activation steering has become a popular way to control Large Language Model (LLM) behavior without fine-tuning. Since the technique is plug-and-play, users share datasets and precomputed vectors
CS-Guard: Benchmarking LLM Guardrails for Code Generation Security
Large language models (LLMs) have been ex- ploited to generate
Expected Harm: Rethinking Safety Evaluation of (Mis)Aligned LLMs
Current evaluations of LLM safety predominantly rely on severity-based taxonomies to assess the harmfulness of malicious queries. We argue that this formulation requires re-examination as it assumes uniform
Understanding Content Moderation in Large Language Models through Restricted Books: From Refusal to Warning
discuss restricted books in only 0.07% of cases, effectively invalidating the premise of jailbreaking research for this content class. Differentiation occurs instead through warning language (+8-15 percentage points