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

203 entries

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

On-Device LLM Hijacking

Published Analyzed Source: arXiv

Mobile LLM agents utilizing vision-based screen perception (OCR or Multimodal Large Language Models) are vulnerable to Visual Prompt Injection via malicious GUI overlays. An attacker holding the SYSTEM_ALERT_WINDOW permission can deploy non-focusable floating windows (using FLAG_NOT_FOCUSABLE) containing adversarial text or fabricated UI elements over legitimate applications. Because the agent captures the entire screen buffer to interpret the device state, it ingests the adversarial overlay…

From Assistants to Adversaries: Exploring the Security Risks of Mobile LLM Agents
Evaluated models: GPT-4o

Judge LLM Prompt Injection

Published Analyzed Source: arXiv

Improper Input Validation in Large Language Model (LLM) systems configured as automated evaluators ("LLM-as-a-judge") allows remote attackers to manipulate evaluation scores and comparative verdicts via adversarial prompt injection. The vulnerability arises when the model processes untrusted input containing linguistic masquerading, context separators, and disruptor commands (e.g., "Basic Injection", "Contextual Misdirection", and "Adaptive Search-Based Attack"). Successful exploitation…

Adversarial Attacks on LLM-as-a-Judge Systems: Insights from Prompt Injections
Evaluated models: GPT-4, Claude 3 Opus, Llama 3.2 3B Instruct+2 more
  • GPT-4
  • Claude 3 Opus
  • Llama 3.2 3B Instruct
  • Gemma 3 4B IT
  • Gemma 3 27B IT

Generative Reward Hacking

Published Analyzed Source: arXiv

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…

Adversarial training of reward models
Evaluated models: Llama 3.1 8B, Llama 3.3 70B, DeepSeek R1+1 more
  • Llama 3.1 8B
  • Llama 3.3 70B
  • DeepSeek R1
  • Gemma 2 27B

Confident Persuasion Override

Published Analyzed Source: arXiv

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…

When persuasion overrides truth in multi-agent llm debates: Introducing a confidence-weighted persuasion override rate (cw-por)
Evaluated models: Llama 3.2 3B, Mistral 7B, Qwen 2.5 14B+1 more
  • Llama 3.2 3B
  • Mistral 7B
  • Qwen 2.5 14B
  • Phi-4 14B

Multi-Agent Prompt Permutation Attack

Published Analyzed Source: arXiv

A vulnerability in multi-agent Large Language Model (LLM) systems allows for a permutation-invariant adversarial prompt attack. By strategically partitioning adversarial prompts and routing them through a network topology, an attacker can bypass distributed safety mechanisms, even those with token bandwidth limitations and asynchronous message delivery. The attack optimizes prompt propagation as a maximum-flow minimum-cost problem, maximizing success while minimizing detection.

Agents Under Siege: Breaking Pragmatic Multi-Agent LLM Systems with Optimized Prompt Attacks
Evaluated models: DeepSeek R1 Distill, Gemma 2 9B, Llama 2 7B+7 more
  • DeepSeek R1 Distill
  • Gemma 2 9B
  • Llama 2 7B
  • Llama 3.1 8B
  • Llama Guard
  • Llama Guard 2 8B
  • Llama Guard 3 1B
  • Llama Guard 3 8B
  • Mistral 7B
  • Prompt Guard

Cross-Batch Interference

Published Analyzed Source: arXiv

Large Language Models (LLMs) deployed using "Batch Prompting" strategies—where multiple distinct user queries are concatenated and processed in a single inference pass to reduce computational costs—are vulnerable to Cross-Query Prompt Injection. When a batch contains a mixture of benign queries and a single malicious query, the instructions within the malicious query (e.g., "apply this rule to every answer") bleed over the context window. This causes the model to apply the adversary's…

Efficient but Vulnerable: Benchmarking and Defending LLM Batch Prompting Attack
Evaluated models: GPT-4o, GPT-4o Mini, Claude 3.5 Sonnet+4 more
  • GPT-4o
  • GPT-4o Mini
  • Claude 3.5 Sonnet
  • Llama 3 70B Instruct
  • Llama 3.2 3B Instruct
  • Qwen 2.5 7B Instruct
  • DeepSeek R1

LLM Fuzz-Based Jailbreak

Published Analyzed Source: arXiv

Large Language Models (LLMs) are vulnerable to jailbreak attacks by crafted prompts that bypass safety mechanisms, causing the model to generate harmful or unethical content. This vulnerability stems from the inherent tension between the LLM's instruction-following and safety constraints. The JBFuzz technique demonstrates the ability to efficiently and effectively discover such prompts through a fuzzing-based approach leveraging novel seed prompt templates and a synonym-based mutation strategy.

JBFuzz: Jailbreaking LLMs Efficiently and Effectively Using Fuzzing
Evaluated models: DeepSeek Chat, DeepSeek R1, Gemini 1.5 Flash+6 more
  • DeepSeek Chat
  • DeepSeek R1
  • Gemini 1.5 Flash
  • Gemini 2.0 Flash
  • GPT-3.5 Turbo
  • GPT-4o
  • GPT-4o Mini
  • Llama 2 7B Chat
  • Llama 3.1 8B Instruct

Life-Cycle Router Misrouting

Published Analyzed Source: arXiv

Predictive Large Language Model (LLM) routers, specifically those utilizing Deep Neural Network (DNN) and Matrix Factorization (MF) architectures, are vulnerable to adversarial manipulation and backdoor poisoning. These routers are designed to optimize cost and latency by dynamically directing simple queries to "weak" (cheap) models and complex queries to "strong" (expensive) models. Attackers can exploit this mechanism in two ways: 1. Inference-time Attacks: By appending specific adversarial…

Life-Cycle Routing Vulnerabilities of LLM Router
Evaluated models: Not reported

LLM Judge Adversarial Vulnerability

Published Analyzed Source: arXiv

Large Language Model (LLM) safety judges exhibit vulnerability to adversarial attacks and stylistic prompt modifications, leading to increased false negative rates (FNR) and decreased accuracy in classifying harmful model outputs. Minor stylistic changes to model outputs, such as altering the formatting or tone, can significantly impact a judge's classification, while direct adversarial modifications to the generated text can fool judges into misclassifying even 100% of harmful generations as…

Know Thy Judge: On the Robustness Meta-Evaluation of LLM Safety Judges
Evaluated models: Atla Selene Mini 8B, Llama 2 13B, Llama 3.1 8B+4 more
  • Atla Selene Mini 8B
  • Llama 2 13B
  • Llama 3.1 8B
  • Llama Guard 3 8B
  • Mistral 7B
  • ShieldGemma 9B
  • WildGuard

Cat-Triggered Reasoning Error

Published Analyzed Source: arXiv

Large Language Models (LLMs) designed for step-by-step problem-solving are vulnerable to query-agnostic adversarial triggers. Appending short, semantically irrelevant text snippets (e.g., "Interesting fact: cats sleep most of their lives") to mathematical problems consistently increases the likelihood of incorrect model outputs without altering the problem's inherent meaning. This vulnerability stems from the models' susceptibility to subtle input manipulations that interfere with their…

Cats Confuse Reasoning LLM: Query Agnostic Adversarial Triggers for Reasoning Models
Evaluated models: DeepSeek R1, DeepSeek R1 Distill Qwen 32B, DeepSeek V3+2 more
  • DeepSeek R1
  • DeepSeek R1 Distill Qwen 32B
  • DeepSeek V3
  • o1
  • o3-mini

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.