Language Model Security Database

Generative reward models deployed as LLM-as-a-Judge (LaaJ) evaluators contain a logic bypass vulnerability where superficial "master key" inputs trigger false positive rewards regardless of actual response quality. Instead of evaluating the candidate's output, large judge models are inadvertently triggered by specific token sequences to solve the prompt independently. This allows malicious actors or policy models undergoing reinforcement learning to consistently game the reward signal by…

A vulnerability in safety-aligned open-weights Large Language Models (LLMs) allows attackers to bypass refusal mechanisms during inference via a representation-level jailbreak known as PCA-regularized Gaussian Optimal Transport (PCA-OT). Unlike previous 1D directional ablation methods (e.g., Refusal Feature Ablation), this attack computes a minimal-cost optimal transport map that matches both the mean and covariance of "harmful" activation distributions to "harmless" ones within a…

Multi-Modal Large Language Models (MLLMs) are vulnerable to a highly transferable, black-box adversarial image attack known as the Multi-Paradigm Collaborative Attack (MPCAttack). Attackers can craft imperceptible visual perturbations by jointly aggregating and optimizing semantic feature representations extracted from surrogate models across three distinct learning paradigms: cross-modal alignment (e.g., CLIP), multi-modal understanding (e.g., InternVL3), and visual self-supervised learning…

A vulnerability in safety-aligned Large Language Models (LLMs) allows attackers to achieve an exponentially scaling Attack Success Rate (ASR) for jailbreaks by combining adversarial prompt injection with repeated inference-time sampling. While ASR against un-injected prompts scales polynomially with the number of generated samples ($k$), introducing a long adversarial suffix acts as a strong "misalignment field." This shifts the model's generation distribution into a replica-symmetric ordered…

Activation steering techniques, such as Contrastive Activation Addition (CAA), systematically erode the safety alignment of Large Language Models (LLMs) due to geometric interference within the residual stream. Steering vectors intended to modulate benign or utility-driven behaviors (e.g., sycophancy, openness, self-awareness) often exhibit a negative cosine similarity with the model's latent 1D refusal direction. When applied during inference, these steering vectors inadvertently suppress the…

Large Language Models (LLMs) are vulnerable to TAO-Attack, an advanced optimization-based jailbreak that bypasses safety alignments by exploiting gradient-guided token updates. The vulnerability stems from a two-stage loss function combined with a Direction-Priority Token Optimization (DPTO) algorithm. In the first stage, the attack optimizes an adversarial prompt suffix to minimize the probability of refusal signals (e.g., "I cannot") while maximizing the probability of a harmful target…

LLaVA-v1.5-7B, when deployed as a vision-language autonomous agent, is highly vulnerable to adversarial image perturbations. An attacker can inject imperceptibly modified images into a web environment (such as an e-commerce storefront). When the VLM agent captures a screenshot containing the perturbed image, the visual noise forces the model to misclassify the scene and output incorrect, structured JSON actions. This allows an attacker to hijack the agent's task execution, bypassing the user's…

The integration of the visual modality in Large Vision-Language Models (VLMs) introduces a vulnerability where appending an image to a harmful text prompt induces a "jailbreak-related representation shift" in the model's internal high-dimensional space. This shift forcibly steers the model's last-token hidden state away from a designated refusal state and into a distinct jailbreak state. The vulnerability occurs because the visual modality overrides the safety alignment of the underlying…

Video-based Large Language Models (Video-LLMs) are vulnerable to a universal Energy-Latency Attack (ELA) that triggers a Denial-of-Service (DoS) via spatially concentrated adversarial patches. Because video architectures rely on temporal subsampling and pooling which act as a low-pass filter against full-frame diffuse noise, an attacker can bypass this compression by anchoring cross-modal attention to a dense, localized visual anomaly. By injecting a fixed, content-agnostic patch into the…

A vulnerability in Vision-Language Models (VLMs) relying on shared visual-textual representation spaces allows attackers to induce transferable cross-task semantic failures using an X-shaped Sparse Pixel Attack (XSPA). Attackers craft imperceptible adversarial perturbations restricted to a fixed geometric prior—two intersecting diagonal lines comprising approximately 1.76% of the image pixels. By jointly optimizing a classification objective with cross-task semantic guidance (target-semantic…