ART optimizes visual pixel inputs to frozen MLLMs to achieve LoRA-competitive accuracy on math and structured tool-use benchmarks without modifying computational graphs.
Visual adversarial examples jailbreak aligned large language models
12 Pith papers cite this work. Polarity classification is still indexing.
12
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
roles
background 2representative citing papers
VisualLeakBench shows vision-language agents propagate PII and unsafe text from images to tool arguments at 78.8% and 85.5% baseline rates, with defensive prompts reducing PII leakage to 2% mainly by suppressing tool use while unsafe-text leakage stays at 52.6%.
A cross-modal alignment attack achieves AUC 0.821 for single-sample black-box membership inference on VLMs such as LLaVA-1.5 by quantifying image-generated caption similarity.
Alignment of vision-language models with human V1-V3 early visual cortex negatively predicts resistance to sycophantic gaslighting attacks.
A narrative survey that catalogs fifty papers on diffusion-based adversarial techniques across text, vision, and vision-language models, proposes a six-class taxonomy of diffusion roles plus a unified five-dimension evaluation framework, and releases a companion catalog.
Introduces MM-Privacy dataset and evaluations showing MLLMs leak sensitive data from images in various tasks, highlighting task inconsistency effects.
FLP uses multi-persona foresight simulation to detect infections via response diversity and applies local purification to reduce maximum cumulative infection rates in multi-agent systems from over 95% to below 5.47%.
Universal adversarial attacks cause output perturbation 90 times more often than precise target injection in VLMs, with only 2 verbatim successes out of 6615 tests.
A multi-agent AI system generates novel biomedical hypotheses that show promising experimental validation in drug repurposing for leukemia, new targets for liver fibrosis, and a bacterial gene transfer mechanism.
MIRAGE creates perceptually benign adversarial images using diffusion and curvature-aware optimization to enable targeted prompt injection attacks on web agents like SeeAct and OpenClaw within attacker-controlled boundaries.
SALLIE detects jailbreaks in text and vision-language models by extracting residual stream activations, scoring maliciousness per layer with k-NN, and ensembling predictions, outperforming baselines on multiple datasets.
The paper proposes a bottom-up framework for safe agentic AI systems that treats each component as a dual-use interface where added capabilities also expand attack surfaces across single agents, multi-agent systems, and interoperable ecosystems.
citing papers explorer
-
Fine-tuning Multi-modal LLMs with ART: Art-based Reinforcement Training
ART optimizes visual pixel inputs to frozen MLLMs to achieve LoRA-competitive accuracy on math and structured tool-use benchmarks without modifying computational graphs.
-
VisualLeakBench: Reproducible Action-Boundary Propagation Failures in Vision-Language Agents
VisualLeakBench shows vision-language agents propagate PII and unsafe text from images to tool arguments at 78.8% and 85.5% baseline rates, with defensive prompts reducing PII leakage to 2% mainly by suppressing tool use while unsafe-text leakage stays at 52.6%.
-
Single-Sample Black-Box Membership Inference Attack against Vision-Language Models via Cross-modal Semantic Alignment
A cross-modal alignment attack achieves AUC 0.821 for single-sample black-box membership inference on VLMs such as LLaVA-1.5 by quantifying image-generated caption similarity.
-
Gaslight, Gatekeep, V1-V3: Early Visual Cortex Alignment Shields Vision-Language Models from Sycophantic Manipulation
Alignment of vision-language models with human V1-V3 early visual cortex negatively predicts resistance to sycophantic gaslighting attacks.
-
Adversarial Diffusion Across Modalities: A Fusion Survey of Attacks, Defenses, and Evaluation for Text, Vision, and Vision-Language Models
A narrative survey that catalogs fifty papers on diffusion-based adversarial techniques across text, vision, and vision-language models, proposes a six-class taxonomy of diffusion roles plus a unified five-dimension evaluation framework, and releases a companion catalog.
-
Unveiling Privacy Risks in Multi-modal Large Language Models: Task-specific Vulnerabilities and Mitigation Challenges
Introduces MM-Privacy dataset and evaluations showing MLLMs leak sensitive data from images in various tasks, highlighting task inconsistency effects.
-
Catching the Infection Before It Spreads: Foresight-Guided Defense in Multi-Agent Systems
FLP uses multi-persona foresight simulation to detect infections via response diversity and applies local purification to reduce maximum cumulative infection rates in multi-agent systems from over 95% to below 5.47%.
-
VisInject: Disruption != Injection -- A Dual-Dimension Evaluation of Universal Adversarial Attacks on Vision-Language Models
Universal adversarial attacks cause output perturbation 90 times more often than precise target injection in VLMs, with only 2 verbatim successes out of 6615 tests.
-
Towards an AI co-scientist
A multi-agent AI system generates novel biomedical hypotheses that show promising experimental validation in drug repurposing for leukemia, new targets for liver fibrosis, and a bacterial gene transfer mechanism.
-
MIRAGE: Stealthy Visual Prompt Injection for Vulnerability Detection in Web Agents
MIRAGE creates perceptually benign adversarial images using diffusion and curvature-aware optimization to enable targeted prompt injection attacks on web agents like SeeAct and OpenClaw within attacker-controlled boundaries.
-
SALLIE: Safeguarding Against Latent Language & Image Exploits
SALLIE detects jailbreaks in text and vision-language models by extracting residual stream activations, scoring maliciousness per layer with k-NN, and ensembling predictions, outperforming baselines on multiple datasets.
-
Toward a Safe Internet of Agents
The paper proposes a bottom-up framework for safe agentic AI systems that treats each component as a dual-use interface where added capabilities also expand attack surfaces across single agents, multi-agent systems, and interoperable ecosystems.