New attack method enhances adversarial transferability in MLLMs

By PulseAugur Editorial · [3 sources] · 2026-05-20 08:15

Researchers have developed FRA-Attack, a novel method to improve the transferability of adversarial attacks against multimodal large language models (MLLMs). This technique utilizes frequency-domain regularization to align perturbations with shared visual cues across different models, overcoming limitations of existing spatial-domain approaches. Experiments on 15 MLLMs demonstrate FRA-Attack's superior performance, particularly against models like GPT-5.4, Claude-Opus-4.6, and Gemini-3-flash. AI

IMPACT Enhances understanding of MLLM vulnerabilities and informs security research.

RANK_REASON The cluster contains an academic paper detailing a new research method.

Read on Hugging Face Daily Papers →

paper
safety

AI-generated summary · Google Gemini · from 3 sources. How we write summaries →

COVERAGE [3]

Hugging Face Daily Papers TIER_1 English(EN) · 2026-05-20 08:15

Frequency-Domain Regularized Adversarial Alignment for Transferable Attacks against Closed-Source MLLMs

Multimodal large language models (MLLMs) remain vulnerable to transfer-based targeted attacks, where perturbations optimized on open-source surrogate encoders can generalize to closed-source MLLMs. A key challenge for improving adversarial transferability is to effectively captur…
arXiv stat.ML TIER_1 English(EN) · Leitao Yuan, Qinghua Mao, Daizong Liu, Kun Wang, Wenjie Wang, Yan Teng, Jing Shao, Dongrui Liu · 2026-05-22 04:00

Frequency-Domain Regularized Adversarial Alignment for Transferable Attacks against Closed-Source MLLMs

arXiv:2605.21541v1 Announce Type: cross Abstract: Multimodal large language models (MLLMs) remain vulnerable to transfer-based targeted attacks, where perturbations optimized on open-source surrogate encoders can generalize to closed-source MLLMs. A key challenge for improving ad…
arXiv stat.ML TIER_1 English(EN) · Dongrui Liu · 2026-05-20 08:15

Frequency-Domain Regularized Adversarial Alignment for Transferable Attacks against Closed-Source MLLMs

Multimodal large language models (MLLMs) remain vulnerable to transfer-based targeted attacks, where perturbations optimized on open-source surrogate encoders can generalize to closed-source MLLMs. A key challenge for improving adversarial transferability is to effectively captur…

COVERAGE [3]

Frequency-Domain Regularized Adversarial Alignment for Transferable Attacks against Closed-Source MLLMs

Frequency-Domain Regularized Adversarial Alignment for Transferable Attacks against Closed-Source MLLMs

Frequency-Domain Regularized Adversarial Alignment for Transferable Attacks against Closed-Source MLLMs

RELATED ENTITIES

RELATED TOPICS