Language Model Security Database

Large Language Models (LLMs) finetuned from open-weight pretrained sources inherit adversarial vulnerabilities encoded in the pretrained model's internal representations. An attacker with white-box access to a pretrained model (e.g., Llama-2, Llama-3) can identify linearly separable features in the hidden states that correlate with "transferable" jailbreak prompts. By exploiting these features using a Probe-Guided Projection (PGP) attack, the attacker can optimize adversarial suffixes on the…

A jailbreak vulnerability, termed Embedded Jailbreak Template (EJT), allows for the generation of harmful content by bypassing the safety mechanisms of Large Language Models (LLMs). The attack uses a generator LLM to contextually integrate a harmful query into a pre-existing jailbreak template. Unlike fixed templates which insert a query into a static placeholder, EJT rewrites multiple parts of the template to embed the harmful intent naturally. This process preserves the original template's…

The KG-DF (Knowledge Graph Defense Framework) contains a logic vulnerability in its Semantic Parsing Module, specifically within the keyword extraction phase defined as $K_{core} = \text{LLM}(P_{prompt})$. The framework relies on a Large Language Model (e.g., GPT-3.5-turbo) to distill user input into keywords ($K_{core}$), which are then embedded to retrieve security warning triples ($T_{match}$) from a Knowledge Graph.

A vulnerability exists in certain Large Language Models and diffusion models due to discontinuities in their latent space, which arise from data sparsity during training. An attacker can craft inputs containing lexically rare or semantically ambiguous constructs to guide the model's inference process toward these unstable, poorly-conditioned regions. This technique, termed "Alignment Degradation Induction," can degrade or bypass safety alignment mechanisms. Through iterative, multi-turn…

Large Language Models (LLMs), specifically open-weights models such as Llama-2, Mistral, and Vicuna, are vulnerable to a white-box adversarial attack framework termed RAID (Refusal-Aware and Integrated Decoding). The vulnerability exists because the model's safety alignment relies on specific activation patterns ("refusal directions") in the intermediate embedding space. Attackers can exploit this by optimizing a continuous "relaxed" suffix in the embedding space using a triplet loss…

Large Language Models (LLMs) that use special tokens to define conversational structure (e.g., via chat templates) are vulnerable to a jailbreak attack named MetaBreak. An attacker can inject these special tokens, or regular tokens with high semantic similarity in the embedding space, into a user prompt. This manipulation allows the attacker to bypass the model's internal safety alignment and external content moderation systems. The attack leverages four primitives: 1. Response Injection…

Large Language Models (LLMs), including Llama 2, Mistral, and Vicuna, are susceptible to a white-box adversarial attack that circumvents safety alignment mechanisms (such as RLHF). The vulnerability exists due to the models' susceptibility to intrinsic optimization of adversarial suffixes using Exponentiated Gradient Descent (EGD). Unlike previous methods that rely on inefficient discrete token searches (e.g., Greedy Coordinate Gradient) or standard projected gradient descent, this attack…

Large Vision-Language Models (LVLMs) that utilize a projection layer (adapter) to bridge a vision encoder and a Large Language Model (LLM) contain a vulnerability stemming from the "Modality Gap"—a distributional distance between image and text token embeddings. This gap allows the visual modality to bypass the safety alignment (RLHF/instruction tuning) of the backbone LLM. Attackers can trigger harmful, toxic, or illegal responses to queries that would be refused in text-only contexts by…

Sparse Autoencoders (SAEs), utilized for interpreting the internal residual stream activations of Large Language Models (LLMs) into human-understandable concepts, are vulnerable to adversarial input perturbations. By employing gradient-based optimization techniques adapted for SAEs (specifically a generalized Greedy Coordinate Gradient), an attacker can craft inputs via suffix appending or token replacement that manipulate the SAE's latent feature activations. This vulnerability allows for the…

A vulnerability exists in Vision-Language Models (VLLMs) that allows for transferable, targeted adversarial attacks. Attackers can generate adversarial image perturbations using an ensemble of open-source surrogate models (primarily CLIP-based visual encoders) which effectively transfer to proprietary, black-box VLLMs. The attack leverages a specific optimization framework that combines a Visual Contrastive Loss with multiple positive/negative visual examples, rather than relying solely on…