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.

Meta-Optimized LLM Judge Jailbreak

Published Analyzed Source: arXiv

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.

Align to Misalign: Automatic LLM Jailbreak with Meta-Optimized LLM Judges
Evaluated models: Claude 3.5 Haiku, Claude 3.5 Sonnet, Claude Sonnet 4+3 more
  • Claude 3.5 Haiku
  • Claude 3.5 Sonnet
  • Claude Sonnet 4
  • GPT-4o
  • GPT-4o Mini
  • Llama 3.1 8B Instruct

Adversarial Self-Deception

Published Analyzed Source: arXiv

Large Language Models (LLMs) exhibit a vulnerability to "adversarial conversational nudges," where the model abandons its internal factual knowledge to align with user-provided misinformation in closed domains (e.g., movies, books). Unlike standard hallucinations where a model lacks knowledge, this vulnerability occurs even when the model demonstrates—via separate self-consistency checks—that it correctly identifies the information as false. When a user creates a multi-turn context asserting…

What About the Scene With the Hitler Reference? HAUNT: A Framework to Probe LLMs' Self-consistency in Closed Domains Via Adversarial Nudge
Evaluated models: GPT-4o, GPT-5, Claude Opus 4+4 more
  • GPT-4o
  • GPT-5
  • Claude Opus 4
  • Gemini 1.5 Flash
  • Gemini 2.5 Flash
  • DeepSeek Reasoner
  • Grok 4

LLM Self-Harm Loop

Published Analyzed Source: arXiv

Large Language Models (LLMs), specifically GPT-3.5-turbo, LLaMA3-8B-instruct, and DeepSeek-R1-Distill-Qwen-7B, are vulnerable to a "Self-Harm" jailbreak attack (Self-HarmLLM). This vulnerability exploits the model's ability to understand its own safety boundaries to generate adversarial inputs against itself. An attacker utilizes a two-session approach: in the first session (Mitigation Session), the attacker instructs the model to rewrite a harmful query into a "Mitigated Harmful Query"…

Self-HarmLLM: Can Large Language Model Harm Itself?
Evaluated models: GPT-3.5 Turbo, Llama 3 8B Instruct, DeepSeek R1 Distill Qwen 7B

Mobile Agent Channel Subversion

Published Analyzed Source: arXiv

Mobile LLM-based agents (including Mobile-Agent-E, AppAgent, AutoDroid, and others) are vulnerable to indirect prompt injection attacks delivered via untrusted third-party mobile channels, such as in-app advertisements, system notifications, and embedded webviews. These agents utilize Multimodal Large Language Models (MLLMs) to perceive the device state via screenshots or accessibility trees. The vulnerability exists because the agents concatenate the user's prompt ($p$) with the environmental…

Measuring the Security of Mobile LLM Agents under Adversarial Prompts from Untrusted Third-Party Channels
Evaluated models: GPT-3.5 Turbo, GPT-4 Turbo, GPT-4o+1 more
  • GPT-3.5 Turbo
  • GPT-4 Turbo
  • GPT-4o
  • GPT-4.1 Mini

Latent Paraphrase Segmentation Attack

Published Analyzed Source: arXiv

Reasoning segmentation models, which generate binary segmentation masks based on implicit text queries, are vulnerable to adversarial paraphrasing. This vulnerability allows an attacker to craft semantically equivalent and grammatically correct text prompts that significantly degrade the model's segmentation performance (measured by Intersection-over-Union, or IoU). The exploit utilizes a black-box, sentence-level optimization method (SPARTA) that operates within the continuous semantic latent…

SPARTA: Evaluating Reasoning Segmentation Robustness through Black-Box Adversarial Paraphrasing in Text Autoencoder Latent Space
Evaluated models: LISA 7B, LISA Explanatory 7B, LISA 13B+3 more
  • LISA 7B
  • LISA Explanatory 7B
  • LISA 13B
  • LISA Explanatory 13B
  • LISA++ 7B
  • GSVA 13B

Internal Prompt Leak Injection

Published Analyzed Source: arXiv

LLM-based coding agents integrated into IDEs (e.g., VS Code Copilot, Cursor, Windsurf) are vulnerable to a query-agnostic Indirect Prompt Injection (IPI) attack termed "QueryIPI." This vulnerability allows an attacker to achieve Remote Code Execution (RCE) on the developer's machine by injecting a malicious tool definition (e.g., via the Model Context Protocol) into the agent's context.

QueryIPI: Query-agnostic Indirect Prompt Injection on Coding Agents
Evaluated models: Claude Sonnet 4

Concurrent Task Jailbreak

Published Analyzed Source: arXiv

A jailbreak vulnerability, known as Task Concurrency, exists in multiple Large Language Models (LLMs). The vulnerability arises when two distinct tasks, one harmful and one benign, are interleaved at the word level within a single prompt. The structure of the malicious prompt alternates words from each task, often using separators like {} to encapsulate words from the second task. This "concurrent" instruction format obfuscates the harmful intent from the model's safety guardrails, causing the…

Adjacent Words, Divergent Intents: Jailbreaking Large Language Models via Task Concurrency
Evaluated models: DeepSeek V3, Gemini 2.5 Flash, GPT-4.1+7 more
  • DeepSeek V3
  • Gemini 2.5 Flash
  • GPT-4.1
  • GPT-4o
  • GPT-4o Mini
  • Llama 2 13B
  • Llama 2 7B
  • Llama 3 8B
  • Mistral 7B
  • Vicuna 13B

Persuasive Jailbreak Fingerprint

Published Analyzed Source: arXiv

Large Language Models (LLMs) are vulnerable to jailbreak attacks that use persuasive techniques grounded in social psychology to bypass safety alignments. Malicious instructions can be reframed using one of Cialdini's seven principles of persuasion (Authority, Reciprocity, Commitment, Social Proof, Liking, Scarcity, and Unity). These rephrased prompts, which remain human-readable and can be generated automatically, manipulate the LLM into complying with harmful requests it would otherwise…

Uncovering the Persuasive Fingerprint of LLMs in Jailbreaking Attacks
Evaluated models: DeepSeek R1, GPT-2, Phi-4+1 more
  • DeepSeek R1
  • GPT-2
  • Phi-4
  • WizardLM Uncensored

LLM Data Instruction Override

Published Analyzed Source: arXiv

Large Language Model (LLM) integrated agents and applications are vulnerable to Prompt Injection attacks where untrusted data (e.g., retrieved documents, tool outputs, website content) overrides system instructions. Because LLMs typically process instructions and data within a single context window without strict separation, an attacker can embed imperative commands within the data channel. This vulnerability extends beyond simple overriding instructions; it includes sophisticated techniques…

Defending against prompt injection with datafilter
Evaluated models: GPT-4o, Llama 3.1 8B Instruct

Adaptive Traversal Jailbreak

Published Analyzed Source: arXiv

Large Language Models (LLMs) including GPT-4o, LLaMA-3, and Mistral-7B are vulnerable to an adaptive multi-turn jailbreak attack known as HarmNet. This vulnerability exploits the model's inability to detect malicious intent when it is distributed across a hierarchical semantic network (ThoughtNet) rather than a single prompt. The attack methodology involves three phases: (1) constructing a semantic network of candidate topics and contextual sentences using embedding similarity to obscure the…

A Framework for Adaptive Multi-Turn Jailbreak Attacks on Large Language Models
Evaluated models: GPT-3.5 Turbo, GPT-4o, Claude 3.5 Sonnet+3 more
  • GPT-3.5 Turbo
  • GPT-4o
  • Claude 3.5 Sonnet
  • Llama 3 8B
  • Mistral 7B
  • Gemma 2 9B

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.