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…

Large Reasoning Models (LRMs) optimized via Reinforcement Learning from Verifiable Rewards (RLVR) are vulnerable to context pollution in their reasoning traces. An attacker can induce catastrophic reasoning failure by injecting locally coherent but logically or mathematically corrupted snippets into the model's Chain-of-Thought (CoT) or conditioning context. Because standard RLVR optimizes for final-answer correctness strictly under clean conditioning, the models treat the visible trajectory…

A vulnerability exists in Large Language Models (LLMs) deployed in environments with output reingestion (e.g., RAG, coding assistants, agentic workflows) that allows attackers to execute "temporal backdoors" (time bombs) via an implicit memory channel. Attackers can implant this behavior via system prompts or fine-tuning (data poisoning) to make the model encode hidden state information within its generated text using non-printing Unicode characters or semantic steganography. When these…

The paper reports a black-box jailbreak evaluation in which a ReAct-style loop adaptively rewrites unsafe text prompts and selectively applies blur, DCT filtering, or recoloring to image regions identified as safety-sensitive. The combined cross-modal strategy is intended to make harmful image-text requests appear less objectionable to a vision-language model while preserving enough semantics to elicit an answer. This is a specific, security-relevant evaluation, although the reported results…

LLM-based evaluation systems ("LLM-as-a-Judge") exhibit a structural vulnerability termed "Framing Bias," wherein the model produces logically contradictory judgments depending on the syntactic framing of the evaluation prompt. Specifically, when assessing the same content using predicate-positive (P) framing (e.g., "Is this toxic?") versus predicate-negative (¬P) framing (e.g., "Is this non-toxic?"), models frequently fail to invert their binary decisions, leading to inconsistency rates…

A vulnerability exists in large language models where safety guardrails can be bypassed by decomposing a single harmful objective into a sequence of individually innocuous sub-queries. An attacker agent can use an adaptive tree search algorithm (Correlated Knowledge Attack Agent - CKA-Agent) to explore the target model's internal correlated knowledge. The agent issues benign queries, uses the model's responses to guide exploration along multiple reasoning paths, and aggregates the collected…

A vulnerability in Large Language Models (LLMs) allows for systematic jailbreaking through a meta-optimization framework called AMIS (Align to MISalign). The attack uses a bi-level optimization process to co-evolve both the jailbreak prompts and the scoring templates used to evaluate them.

Pattern Enhanced Chain of Attack (PE-CoA) shows that safety controls can be bypassed gradually across a conversation. The paper evaluates five recurring conversational patterns and finds that resistance to one pattern does not reliably generalize to others, creating a black-box, multi-turn jailbreak risk even when individual turns appear benign.

A vulnerability exists in multiple Large Language Models (LLMs) where an attacker can bypass safety alignments by exploiting the model's ethical reasoning capabilities. The attack, named TRIAL (Trolley-problem Reasoning for Interactive Attack Logic), frames a harmful request within a multi-turn ethical dilemma modeled on the trolley problem. The harmful action is presented as the "lesser of two evils" necessary to prevent a catastrophic outcome, compelling the model to engage in utilitarian…

AdvEDM reveals a vulnerability in Vision-Language Model (VLM) based Embodied Decision-Making (EDM) systems, such as those used in autonomous driving and robotic manipulation. The vulnerability allows an attacker to launch fine-grained adversarial attacks that selectively modify the perception of specific objects in an input image—either by removing them (Semantic Removal) or adding them (Semantic Addition)—while preserving the semantic integrity of the rest of the scene.