Language Model Security Database

A vulnerability exists in multiple LLMs allowing attackers to elicit harmful responses by strategically distributing malicious intent across multiple turns in a conversation. The vulnerability is not detected by single-turn safety measures, as the harmful intent is only revealed through a sequence of seemingly benign prompts. The vulnerability is exacerbated by the use of techniques such as prompt optimization that dynamically adjust prompts based on model responses, maximizing the likelihood…

Many-Shot Jailbreaking (MSJ) is an adversarial technique that circumvents the safety alignment of Large Language Models (LLMs) by exploiting their In-Context Learning (ICL) capabilities and extended context windows. By embedding a large number of "shots" (fake dialogue examples) within a single prompt—where a simulated assistant complies with harmful requests—the attacker conditions the model to ignore its safety training. As the number of malicious examples increases (following a power-law…

A vulnerability in Large Language Models (LLMs) allows attackers to bypass safety mechanisms and elicit detailed harmful responses by strategically manipulating input prompts. The vulnerability exploits the LLM's sensitivity to "scenario shifts"—contextual changes in the input that influence the model's output, even when the core malicious request remains the same. A genetic algorithm can optimize these scenario shifts, increasing the likelihood of obtaining detailed harmful responses while…

Large Language Models (LLMs) are vulnerable to a jailbreaking attack leveraging humorous prompts. Embedding an unsafe request within a humorous context, using a fixed template, bypasses built-in safety mechanisms and elicits unsafe responses. The attack's success relies on a balance; too little or too much humor reduces effectiveness.

Large Language Models (LLMs) exhibit Defense Threshold Decay (DTD): generating substantial benign content shifts the model's attention from the input prompt to prior outputs, increasing susceptibility to jailbreak attacks. The "Sugar-Coated Poison" (SCP) attack exploits this by first generating benign content, then transitioning to malicious output.

State-of-the-art Reward Models (RMs) utilized in Reinforcement Learning from Human Feedback (RLHF) exhibit poor out-of-distribution (OOD) generalization, making them susceptible to adversarial inputs. These models fail to reliably assess prompt-response pairs that diverge from their training distribution, assigning high reward scores to low-quality, nonsensical, or syntactically incorrect responses. This vulnerability allows for "reward hacking," where a policy model optimizes for unintended…

Retrieval-Augmented Generation (RAG) systems are vulnerable to a targeted corpus poisoning attack known as "CorruptRAG". This vulnerability allows an attacker to manipulate the response of an LLM to a specific target query by injecting a single malicious document into the RAG knowledge database. Unlike traditional poisoning attacks that require flooding the retrieval results (top-N) with malicious content to outnumber correct information, CorruptRAG succeeds with a single retrieved document.

Multimodal Large Language Models (MLLMs) are vulnerable to a jailbreaking attack, dubbed PiCo, that leverages token-level typographic attacks on images embedded within code-style instructions. The attack bypasses multi-tiered defense mechanisms, including input filtering and runtime monitoring, by exploiting weaknesses in the visual modality's integration with programming contexts. Harmful intent is concealed within visually benign image fragments and code instructions, circumventing safety…

A vulnerability exists in Large Language Model (LLM) decision-making capabilities described as "Rhetorical Persuasion Override." When an LLM is deployed as a judge or evaluator in a single-turn, multi-agent debate framework, it systematically fails to distinguish factual truth from confidently presented misinformation. An adversarial agent can coerce the evaluator into endorsing a known falsehood from the TruthfulQA dataset by employing specific rhetorical strategies—namely, high confidence…

Multilingual and multi-accent audio inputs, combined with acoustic adversarial perturbations (reverberation, echo, whisper effects), can bypass safety mechanisms in Large Audio Language Models (LALMs), causing them to generate unsafe or harmful outputs. The vulnerability is amplified by the interaction between acoustic and linguistic variations, particularly in languages with less training data.