Language Model Security Database

Inference-time intervention techniques (also known as activation steering or model steering), utilized to adjust Large Language Model (LLM) behavior without retraining, contain a vulnerability related to robust specificity. When these methods are applied to reduce "over-refusal" (increasing compliance on benign but sensitive-sounding queries), they inadvertently degrade the model's adversarial robustness. Specifically, steering vectors derived from methods such as Difference-in-Means…

A vulnerability exists in aligned Large Language Models (LLMs) related to "shallow safety alignment," where safety mechanisms disproportionately rely on the initial tokens generated by the model. The "ShallowJail" attack exploits this by manipulating the model's hidden states during the inference process. Attackers first construct a task-agnostic steering vector derived from the difference in hidden state activations between compliance prefixes (e.g., "Sure, here are the details") and refusal…

Large Language Models (LLMs) are vulnerable to jailbreak attacks that exploit the positional sensitivity of adversarial tokens. Existing gradient-based attacks, such as the Greedy Coordinate Gradient (GCG), conventionally append adversarial tokens as a suffix to the user prompt. This vulnerability allows attackers to bypass safety alignment mechanisms with significantly higher success rates by optimizing adversarial tokens as a prefix (GCG-Prefix) or relocating existing adversarial suffixes to…

Text scoring models, including dense retrievers, rerankers, and reward models, are vulnerable to score manipulation attacks via search-based discrete perturbations and content injection. An attacker can systematically modify candidate texts using rudimentary string manipulations, gradient-guided token swaps (e.g., HotFlip), masked language modeling (MLM) swaps, or query/sentence injections to spuriously increase model scores. This structural failure condition allows an irrelevant passage or a…

Vision Language Models (VLMs) utilizing independent vision encoders (e.g., ViT) and Large Language Model (LLM) decoders are vulnerable to Split-Image Visual Jailbreak Attacks (SIVA). The vulnerability arises from an architectural and alignment discrepancy: while the vision encoder processes image fragments (splits) in isolation via constrained attention or block-diagonal masks, the LLM decoder aggregates these features via cross-attention to reconstruct the semantic content. Current safety…

Large Vision-Language Models (LVLMs) are vulnerable to Visual Memory Injection (VMI), a stealthy targeted attack targeting multi-turn conversations. An attacker can embed an imperceptible adversarial perturbation ($L_\infty \le 8/255$) into a seemingly benign image. Because the visual input persists in the model's context throughout a multi-turn dialogue, the injected payload remains dormant. By utilizing "benign anchoring" and "context-cycling" during optimization, the attacker ensures the…

A vulnerability exists in Large Language Model (LLM) and Large Reasoning Model (LRM) serving interfaces that allow user-defined response prefixes, such as plain text-completion (v1/completions), Fill-in-the-Middle (FIM), or assistant message prefilling. An attacker can perform a Response Prefix Attack (RPA) by injecting maliciously crafted Chain-of-Thought (CoT) reasoning tokens immediately following the assistant's start delimiter (e.g., <|im_start|>assistant). Because these tokens are placed…

Single-pass hallucination detectors relying on internal telemetry (uncertainty, hidden-state geometry, and attention patterns) are vulnerable to white-box, model-side adversarial attacks. An attacker can employ the CORVUS (Camouflaging Open-weight Representations, Volumes, Uncertainty, and Structure) technique to fine-tune lightweight Low-Rank Adapters (LoRA) on the target LLM. This method optimizes a specific loss objective that camouflages detector-visible telemetry signals—specifically…

A vulnerability exists in aligned Large Language Models (LLMs) where inducing "drunk language" behavior—simulating the text of an intoxicated human—bypasses safety guardrails and contextual privacy protections. Attackers can exploit this anthropomorphic flaw through inference-time persona prompting or lightweight post-training (causal fine-tuning or reinforcement learning on drunk text corpora). By forcing the model to adopt a stylistic and semantic framework associated with impaired human…

A safety bypass vulnerability exists in aligned Large Language Models (LLMs) permitting inference-time jailbreaking via direct activation repatching. The vulnerability exploits the distributed nature of safety mechanisms, which are governed by approximately 30% of total attention heads (termed "safety-critical heads"). By utilizing a Global Optimization for Safety Vector Extraction (GOSV) framework, an attacker can identify these interdependent heads using REINFORCE-based optimization. Once…