REVIEW 19 cited by
Image Hijacks: Adversarial Images can Control Generative Models at Runtime
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Image Hijacks: Adversarial Images can Control Generative Models at Runtime
read the original abstract
Are foundation models secure against malicious actors? In this work, we focus on the image input to a vision-language model (VLM). We discover image hijacks, adversarial images that control the behaviour of VLMs at inference time, and introduce the general Behaviour Matching algorithm for training image hijacks. From this, we derive the Prompt Matching method, allowing us to train hijacks matching the behaviour of an arbitrary user-defined text prompt (e.g. 'the Eiffel Tower is now located in Rome') using a generic, off-the-shelf dataset unrelated to our choice of prompt. We use Behaviour Matching to craft hijacks for four types of attack, forcing VLMs to generate outputs of the adversary's choice, leak information from their context window, override their safety training, and believe false statements. We study these attacks against LLaVA, a state-of-the-art VLM based on CLIP and LLaMA-2, and find that all attack types achieve a success rate of over 80%. Moreover, our attacks are automated and require only small image perturbations.
Forward citations
Cited by 19 Pith papers
-
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 ...
-
Catching the Infection Before It Spreads: Foresight-Guided Defense in Multi-Agent Systems
A foresight-based local purification method using multi-persona simulations and recursive diagnosis reduces infectious jailbreak spread in multi-agent systems from over 95% to below 5.47% while matching benign perform...
-
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.
-
Visual Adversarial Attack on Vision-Language Models for Autonomous Driving
ADvLM is the first visual adversarial attack framework for VLMs in autonomous driving, using semantic-invariant induction via LLM-generated prompt libraries and scenario-associated attention-based enhancement to achie...
-
On Adversarial Vulnerability of Vision-Language Models through the Lens of Intermediate Spectral Subspaces
Aligning adversarial perturbations with the near-null singular directions of intermediate linear layers in transformer VLMs yields stronger attacks than existing feature- and output-space methods.
-
Exploring Adversarial Robustness and Safety Alignment in Multilingual Multi-Modal Large Language Models
Adversarial images transfer across languages in MLLMs while apparent safety in weaker languages stems from comprehension and visual-grounding failures rather than genuine alignment.
-
Benign Inputs, Harmful Outputs: Cross-Modal Jailbreaking via Distributed Semantic Recomposition
DSR decomposes harmful intents into benign textual and visual primitives that MLLMs fuse into harmful outputs, achieving high attack success with low input toxicity.
-
Guaranteed Jailbreaking Defense via Disrupt-and-Rectify Smoothing
DR-Smoothing introduces a disrupt-then-rectify prompt processing scheme into smoothing defenses, delivering tight theoretical bounds on success probability against both token- and prompt-level jailbreaks.
-
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%.
-
Catching the Infection Before It Spreads: Foresight-Guided Defense in Multi-Agent Systems
A foresight-based local purification method simulates future agent interactions, detects infections via response diversity across personas, and applies targeted rollback or recursive diagnosis to cut maximum infection...
-
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.
-
RedDiffuser: Auditing Multimodal Safety Failures in Vision-Language Models via Reinforced Diffusion
RedDiffuser is a reinforced diffusion framework that generates adversarial visual contexts to audit and expose widespread multimodal safety failures in VLMs, increasing unsafe response rates by up to 10.69% on LLaVA w...
-
A StrongREJECT for Empty Jailbreaks
StrongREJECT provides a standardized benchmark and evaluator for jailbreak attacks that aligns better with human judgments than prior methods and reveals that successful jailbreaks often reduce model capabilities.
-
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.
-
Laundering AI Authority with Adversarial Examples
Adversarial examples enable AI authority laundering by causing production VLMs to give authoritative but wrong responses on subtly perturbed images, with success rates of 22-100% using decade-old attack methods.
-
Surviving the Unseen: Predictive Defense for Novel Multi-Turn Multimodal Attacks
Proposes the TRIAD framework that treats multi-turn multimodal attacks as continuous trajectories and uses structural anomaly detection, regularized Mahalanobis distance, topological acceleration, and a time-varying C...
-
AI Safety Landscape for Large Language Models: Taxonomy, State-of-the-art, and Future Directions
The paper introduces a taxonomy of AI safety for LLMs organized into Trustworthy AI, Responsible AI, and Safe AI perspectives, accompanied by a review of state-of-the-art methods, challenges, and future directions.
-
Safety at Scale: A Comprehensive Survey of Large Model and Agent Safety
A comprehensive survey that taxonomizes safety threats to large models and agents, reviews defenses and benchmarks, and outlines open challenges.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.