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

468 entries

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

X-Shaped Sparse VLM Attack

Published Analyzed Source: arXiv

A vulnerability in Vision-Language Models (VLMs) relying on shared visual-textual representation spaces allows attackers to induce transferable cross-task semantic failures using an X-shaped Sparse Pixel Attack (XSPA). Attackers craft imperceptible adversarial perturbations restricted to a fixed geometric prior—two intersecting diagonal lines comprising approximately 1.76% of the image pixels. By jointly optimizing a classification objective with cross-task semantic guidance (target-semantic…

XSPA: Crafting Imperceptible X-Shaped Sparse Adversarial Perturbations for Transferable Attacks on VLMs
Evaluated models: InstructBLIP

Zero-Tax Fine-Tune Bypass

Published Analyzed Source: arXiv

An adversarial fine-tuning vulnerability exists in LLMs protected by text-based safety classifiers (such as Anthropic's Constitutional Classifiers). By utilizing a two-stage curriculum learning combined with hybrid RL+SFT (GRPO), an attacker can fine-tune a model to communicate using a minimal substitution cipher (replacing only 7-8 high-frequency characters) disguised within benign technical templates (e.g., forensic logs with 0x prefixes). This "Trojan-Speak" methodology bypasses text-level…

Trojan-Speak: Bypassing Constitutional Classifiers with No Jailbreak Tax via Adversarial Finetuning
Evaluated models: Claude Haiku 4.5, Qwen 3 4B, Qwen 3 8B+2 more
  • Claude Haiku 4.5
  • Qwen 3 4B
  • Qwen 3 8B
  • Qwen 3 14B
  • Qwen 3 32B

Autoresearch Adversarial Algorithm Discovery

Published Analyzed Source: arXiv

GPT-OSS-Safeguard-20B and Meta-SecAlign (70B/8B) are vulnerable to white-box adversarial attacks generated by automated algorithmic recombination (specifically the claude_v63, claude_v82, and claude_v53-oss optimizers). These algorithms significantly outperform standard discrete optimization methods (like GCG) by integrating continuous optimization (ADC) with LayerNorm gradient scaling (LSGM), or by merging momentum-smoothed gradients with directional perturbation candidate selection (DPTO)…

Claudini: Autoresearch Discovers State-of-the-Art Adversarial Attack Algorithms for LLMs
Evaluated models: Llama 2 7B, Llama 3 8B, Qwen 2.5 7B+2 more
  • Llama 2 7B
  • Llama 3 8B
  • Qwen 2.5 7B
  • Gemma 2 2B
  • Gemma 7B

Steering Vector Safety Gap

Published Analyzed Source: arXiv

Activation steering techniques, such as Contrastive Activation Addition (CAA), systematically erode the safety alignment of Large Language Models (LLMs) due to geometric interference within the residual stream. Steering vectors intended to modulate benign or utility-driven behaviors (e.g., sycophancy, openness, self-awareness) often exhibit a negative cosine similarity with the model's latent 1D refusal direction. When applied during inference, these steering vectors inadvertently suppress the…

Analysing the Safety Pitfalls of Steering Vectors
Evaluated models: Llama 2 7B, Qwen 2.5 3B, Gemma 7B

LLM Universal Graph Subversion

Published Analyzed Source: arXiv

An evasion vulnerability in Text-Attributed Graph (TAG) learning models allows attackers to induce targeted misclassifications via LLM-generated, coordinated perturbations to both graph topology and textual semantics. By identifying a semantically distant "influencer" node, an attacker can use a separate LLM to selectively delete highly relevant edges, insert a deceptive edge connecting the target to the influencer, and slightly modify the target node's text to include a keyword aligned with…

Can LLMs Fool Graph Learning? Exploring Universal Adversarial Attacks on Text-Attributed Graphs
Evaluated models: DeepSeek-V3 671B, Llama 4 17B, Mistral 7B+1 more
  • DeepSeek-V3 671B
  • Llama 4 17B
  • Mistral 7B
  • Qwen Plus

Cross-Language Safety Drift

Published Analyzed Source: arXiv

Leading Large Language Models (LLMs) exhibit significant cross-lingual safety drift, allowing users to bypass safety guardrails by translating harmful prompts into low-resource Indic languages. While models effectively block unsafe prompts concerning caste, religion, gender, and politics in high-resource languages like English and Hindi, their safety alignment severely degrades in low-resource scripts such as Odia, Telugu, Kannada, and Punjabi. Evaluated models demonstrate a cross-language…

IndicSafe: A Benchmark for Evaluating Multilingual LLM Safety in South Asia
Evaluated models: GPT-4o Mini, Claude Sonnet 4, Grok 3+6 more
  • GPT-4o Mini
  • Claude Sonnet 4
  • Grok 3
  • Llama 3.3 70B Instruct
  • Llama 3.1 405B
  • Qwen 1.5 7B Chat
  • Mistral 7B Instruct v0.2
  • Command R
  • Command A

VLM Image-Shift Jailbreak

Published Analyzed Source: arXiv

The integration of the visual modality in Large Vision-Language Models (VLMs) introduces a vulnerability where appending an image to a harmful text prompt induces a "jailbreak-related representation shift" in the model's internal high-dimensional space. This shift forcibly steers the model's last-token hidden state away from a designated refusal state and into a distinct jailbreak state. The vulnerability occurs because the visual modality overrides the safety alignment of the underlying…

Understanding and Defending VLM Jailbreaks via Jailbreak-Related Representation Shift
Evaluated models: LLaVA 1.5 7B, ShareGPT4V 7B, InternVL-Chat 19B

VLM E-commerce Attack Surface

Published Analyzed Source: arXiv

LLaVA-v1.5-7B, when deployed as a vision-language autonomous agent, is highly vulnerable to adversarial image perturbations. An attacker can inject imperceptibly modified images into a web environment (such as an e-commerce storefront). When the VLM agent captures a screenshot containing the perturbed image, the visual noise forces the model to misclassify the scene and output incorrect, structured JSON actions. This allows an attacker to hijack the agent's task execution, bypassing the user's…

Adversarial attacks against Modern Vision-Language Models
Evaluated models: Qwen 2.5 VL 7B Instruct, LLaVA 1.5 7B

Contextual Priority Hacking

Published Analyzed Source: arXiv

Large Language Models (LLMs) are vulnerable to a jailbreak technique termed "Priority Hacking." Adversaries can bypass safety alignments by exploiting the model's internal priority graph, where certain abstract values (e.g., justice, public health) implicitly outweigh general safety restrictions within specific contexts. By crafting a deceptive prompt that frames a malicious request as a necessary action in service of a higher-priority benign value, attackers engineer a value conflict. The…

Are Dilemmas and Conflicts in LLM Alignment Solvable? A View from Priority Graph
Evaluated models: Not reported

Visual Confused Deputy Attack

Published Analyzed Source: arXiv

A "Visual Confused Deputy" vulnerability exists in Computer-Using Agents (CUAs) that rely on visual perception to execute coordinate-based GUI actions (e.g., click(x,y)). Because the agent's understanding of the system state is entirely dependent on the screenshot provided by the runtime, a compromised runtime or tool can intercept and alter the screenshot pixels before forwarding them to the LLM. By visually swapping the locations of benign and privileged UI elements, an attacker can trick…

Visual Confused Deputy: Exploiting and Defending Perception Failures in Computer-Using Agents
Evaluated models: Claude 3.7 Sonnet

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.