Skip to main content
LLM Security Database
Skip to research search
Last analyzed Sep 9, 2026

Language Model Security Database

985 research findings · 1119 evaluated models

Filtered research findings

610 entries

Matches every word across titles, descriptions, sources, affected systems, and models.

Simulated Ensemble Jailbreak Transfer

Published Analyzed Source: arXiv

A vulnerability exists in the fine-tuning lifecycle of Vision-Language Models (VLMs) derived from open-source base models, termed the "grey-box threat." Adversaries with white-box access to a public base model (e.g., Qwen2-VL) can generate universal adversarial images that successfully bypass safety guardrails in proprietary, fine-tuned downstream variants. This is achieved via the Simulated Ensemble Attack (SEA), which combines two techniques: Fine-tuning Trajectory Simulation (FTS), where…

Simulated Ensemble Attack: Transferring Jailbreaks Across Fine-tuned Vision-Language Models
Evaluated models: Qwen 2 2B

Word Puzzle Reasoning Jailbreak

Published Analyzed Source: arXiv

A logic-based jailbreak vulnerability exists in Large Language Models (LLMs) known as "PUZZLED," where safety alignment mechanisms are bypassed by embedding harmful instructions within word-based puzzles. The attacker identifies sensitive keywords in a malicious prompt, masks them (e.g., replacing "bomb" with "[WORD1]"), and presents the masked terms as a cognitive task—specifically Word Searches, Anagrams, or Crosswords—accompanied by linguistic clues (word length, part-of-speech, and…

PUZZLED: Jailbreaking LLMs through Word-Based Puzzles
Evaluated models: GPT-4.1, GPT-4o, Claude 3.7 Sonnet+2 more
  • GPT-4.1
  • GPT-4o
  • Claude 3.7 Sonnet
  • Gemini 2.0 Flash
  • Llama 3.1 8B Instruct

Activation-Guided Local Editing Jailbreak

Published Analyzed Source: arXiv

A vulnerability exists in multiple Large Language Models (LLMs) that allows for safety alignment bypass through a technique named Activation-Guided Local Editing (AGILE). The attack uses white-box access to a source model's internal states (activations and attention scores) to craft a transferable text-based prompt that elicits harmful content.

Activation-Guided Local Editing for Jailbreaking Attacks
Evaluated models: Claude 3.5 Sonnet, DarkIdol Llama 3.1 8B Instruct, DeepSeek V3+9 more
  • Claude 3.5 Sonnet
  • DarkIdol Llama 3.1 8B Instruct
  • DeepSeek V3
  • Gemini 2.0 Flash
  • GLM 4 9B Chat
  • GPT-4o
  • Llama 2 7B Chat
  • Llama 3 8B Instruct
  • Llama 3.1 8B Instruct
  • Llama 3.2 3B Instruct
  • Phi-4 Mini Instruct
  • Qwen 2.5 7B Instruct

Full-Spectrum Diffusion Attack

Published Analyzed Source: arXiv

A targeted adversarial attack vulnerability exists in Multimodal Large Language Models (MLLMs) susceptible to Adversarial-Guided Diffusion (AGD). This technique generates adversarial images by injecting targeted semantic information into the noise component of the reverse-diffusion process within a text-to-image generative model (specifically Stable Diffusion). Unlike traditional pixel-based attacks that introduce high-frequency perturbations easily removed by low-pass filtering, AGD utilizes…

Adversarial-guided diffusion for multimodal llm attacks
Evaluated models: Vicuna 13B

Synergistic Bias Jailbreak

Published Analyzed Source: arXiv

Large Language Models (LLMs) utilizing Reinforcement Learning from Human Feedback (RLHF) and other safety alignment techniques are vulnerable to "CognitiveAttack," a jailbreak vector that exploits synergistic cognitive biases. The vulnerability exists because models internalize human-like reasoning fallacies during pre-training and alignment. Adversaries can bypass safety guardrails by rewriting harmful instructions to trigger specific psychological heuristics—specifically through the…

Exploiting Synergistic Cognitive Biases to Bypass Safety in LLMs
Evaluated models: GPT-3.5, Llama 2 7B, DeepSeek R1+2 more
  • GPT-3.5
  • Llama 2 7B
  • DeepSeek R1
  • DeepSeek V3
  • o4-mini

Prompt-Based Jailbreak Taxonomy

Published Analyzed Source: arXiv

Large Language Models (LLMs) and Text-to-Image (T2I) models are vulnerable to jailbreaking through prompt-based attacks that use narrative framing, semantic substitution, and context diffusion to bypass safety moderation pipelines. These attacks do not require specialized knowledge or technical expertise. Attackers can embed harmful requests within benign narratives, frame them as fictional or professional inquiries, or use euphemistic language to circumvent input filters and output…

Anyone Can Jailbreak: Prompt-Based Attacks on LLMs and T2Is
Evaluated models: Not reported

Agent Policy Hacking

Published Analyzed Source: arXiv

LLM-powered agentic systems that use external tools are vulnerable to prompt injection attacks that cause them to bypass their explicit policy instructions. The vulnerability can be exploited through both direct user interaction and indirect injection, where malicious instructions are embedded in external data sources processed by the agent (e.g., documents, API responses, webpages). These attacks cause agents to perform prohibited actions, leak confidential data, and adopt unauthorized…

Security challenges in ai agent deployment: Insights from a large scale public competition
Evaluated models: Claude 3.5 Sonnet, Claude 3.7 Sonnet, Command R+11 more
  • Claude 3.5 Sonnet
  • Claude 3.7 Sonnet
  • Command R
  • Command R+
  • Gemini 1.5 Flash
  • Gemini 1.5 Pro
  • Gemini 2.0 Flash
  • Gemini 2.5 Pro
  • GPT-4.5
  • GPT-4o
  • Llama 3.3 70B
  • o3
  • o3-mini
  • o4-mini

Persona-Enhanced Genetic Jailbreak

Published Analyzed Source: arXiv

A vulnerability exists where Large Language Models (LLMs) can be manipulated by prepending a specially crafted 'persona prompt', often in the system prompt. These persona prompts cause the model to shift its attention from sensitive keywords in a harmful request to the stylistic instructions of the persona. This weakens the model's safety alignment, significantly reducing its refusal rate for harmful queries. The vulnerability is particularly severe because these persona prompts have a…

Enhancing Jailbreak Attacks on LLMs via Persona Prompts
Evaluated models: DeepSeek V3, GPT-4o, GPT-4o Mini+2 more
  • DeepSeek V3
  • GPT-4o
  • GPT-4o Mini
  • Llama 3.1 8B Instruct
  • Qwen 2.5 14B Instruct

LLM Interpreter Resource Exhaustion

Published Analyzed Source: arXiv

Large Language Models (LLMs) equipped with native code interpreters are vulnerable to Denial of Service (DoS) via resource exhaustion. An attacker can craft a single prompt that causes the interpreter to execute code that depletes CPU, memory, or disk resources. The vulnerability is particularly pronounced when a resource-intensive task is framed within a plausibly benign or socially-engineered context ("indirect prompts"), which significantly lowers the model's likelihood of refusal compared…

Running in CIRCLE? A Simple Benchmark for LLM Code Interpreter Security
Evaluated models: Gemini 2.0 Flash, Gemini 2.5 Flash, Gemini 2.5 Pro+5 more
  • Gemini 2.0 Flash
  • Gemini 2.5 Flash
  • Gemini 2.5 Pro
  • GPT-4.1
  • GPT-4.1 Mini
  • GPT-4.1 Nano
  • o3 Pro
  • o4-mini

LVLM Vision-Triggered Resource Exhaustion

Published Analyzed Source: arXiv

A resource consumption vulnerability exists in multiple Large Vision-Language Models (LVLMs). An attacker can craft a subtle, imperceptible adversarial perturbation and apply it to an input image. When this image is processed by an LVLM, even with a benign text prompt, it forces the model into an unbounded generation loop. The attack, named RECALLED, uses a gradient-based optimization process to create a visual perturbation that steers the model's text generation towards a predefined…

Resource Consumption Red-Teaming for Large Vision-Language Models
Evaluated models: LLaVA 1.5 7B, LLaVA 1.5 13B, Qwen 2.5 VL 3B Instruct+4 more
  • LLaVA 1.5 7B
  • LLaVA 1.5 13B
  • Qwen 2.5 VL 3B Instruct
  • Qwen 2.5 VL 7B Instruct
  • Qwen 2.5 VL 32B Instruct
  • InstructBLIP Vicuna 7B
  • InstructBLIP Vicuna 13B

Research methodology

Entries summarize publicly available primary-source security research. Model names reflect only systems explicitly evaluated by the cited paper, and measurements are research-reported unless independent verification is stated.

Discovery and publication are separate: new papers enter a candidate queue, then require human review and explicit approval before publication. Updates follow that review process, not a guaranteed daily publishing schedule. “Last analyzed” is the newest analysis date among published entries; it does not mean that all recent research has been covered.