StructBreak: Structural Cognitive Overload-Induced Safety Failures in MLLMs

• URL: http://arxiv.org/abs/2605.25534v1
• Date: Mon, 25 May 2026 07:41:51 GMT
• Title: StructBreak: Structural Cognitive Overload-Induced Safety Failures in MLLMs
• Abstract summary: Multimodal Large Language Models (MLLMs) excel at structural reasoning yet suffer from a sharp logical brittleness in structural consistency.<n>We term this phenomenon Structural Cognitive Overload (SCO), a byproduct of the contention between deep reasoning and safety alignment.<n>We propose StructBreak, an automated end-to-end framework designed to quantify SCO.
• Score: 20.638038966630337
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Multimodal Large Language Models (MLLMs) excel at structural reasoning yet suffer from a sharp logical brittleness in structural consistency. We term this phenomenon Structural Cognitive Overload (SCO), a byproduct of the contention between deep reasoning and safety alignment. However, prior work has predominantly targeted typographic and pixel-level perturbations, leaving the study of SCO largely unexplored. To this end, we propose StructBreak, an automated end-to-end framework designed to quantify SCO. By leveraging StructBreak, we uncover a novel higher-order cognitive overload attack paradigm; notably, this attack operates under a practical black-box setting, requiring no internal model access. Consequently, we utilize this framework to establish a comprehensive benchmark spanning ten diverse threat scenarios. Empirical evaluations on six leading MLLMs reveal that SCO readily triggers toxic generation, yielding a 92% average ASR (up to 97% on Gemini 2.5). To elucidate the mechanism of SCO, we further conduct model-level interpretations spanning attention dynamics, latent space topology, and geometric analysis. Our findings reveal that StructBreak acts as a novel structural channel to circumvent safety filters. Furthermore, the limited efficacy of inherent safety mechanisms underscores that current alignment paradigms are insufficient for the era of complex multimodal reasoning.

Related papers

• Provably Secure Agent Guardrail [89.79561918065122]
  Existing defense architectures rely on empirical semantic guardrails and probabilistic large model adjudicators.<n>This paper proposes a new security paradigm for agents based on the fundamental limitations of logical reasoning.
  arXiv  Detail & Related papers  (2026-05-28T02:12:41Z)
• When Interpretability Becomes a Liability: Adversarial Attacks on CBM Concept Layers [0.0]
  Concept Bottleneck Models (CBMs) have emerged as a cornerstone approach for interpretable machine learning.<n>We present a comprehensive, systematic study of concept-level adversarial vulnerabilities in CBMs.<n>We introduce SPECTRA, a stability principled regularization defense.
  arXiv  Detail & Related papers  (2026-05-25T00:03:47Z)
• Reasoning as an Attack Surface: Adaptive Evolutionary CoT Jailbreaks for LLMs [56.53007443197966]
  Large Reasoning Models (LRMs) have demonstrated remarkable capabilities in reasoning and generation tasks.<n>Their explicit chain-of-thought (CoT) mechanism introduces new security risks, making them particularly vulnerable to jailbreak attacks.<n>We propose an adaptive evolutionary CoT jailbreak framework, called AE-CoT, to overcome these limitations.
  arXiv  Detail & Related papers  (2026-05-23T10:11:32Z)
• METER: Evaluating Multi-Level Contextual Causal Reasoning in Large Language Models [61.33372454250959]
  Contextual causal reasoning is a critical yet challenging capability for Large Language Models.<n>Existing benchmarks fail to ensure context consistency or cover the full causal hierarchy.<n>We pioneer METER to systematically benchmark LLMs across all three levels of the causal ladder.
  arXiv  Detail & Related papers  (2026-04-13T14:07:11Z)
• From Intent to Evidence: A Categorical Approach for Structural Evaluation of Deep Research Agents [25.080554451314683]
  Deep research agents (DRAs) have emerged as a promising paradigm for complex information synthesis.<n>We formalize DRA behavior through the lens of category theory, modeling deep research workflow as a composition of structure-preserving maps.<n>We introduce a novel mechanism-aware benchmark with 296 questions designed to stress-test agents along four interpretable axes.
  arXiv  Detail & Related papers  (2026-03-26T11:37:26Z)
• RooflineBench: A Benchmarking Framework for On-Device LLMs via Roofline Analysis [53.90240071275054]
  The transition toward localized intelligence through Small Language Models (SLMs) has intensified the need for rigorous performance characterization on resource-constrained edge hardware.<n>We propose a systematic framework that unifies architectural primitives and hardware constraints through the lens of operational intensity (OI)<n>By defining an inference-potential region, we introduce the Relative Inference Potential as a novel metric to compare efficiency differences between Large Language Models (LLMs) on the same hardware substrate.
  arXiv  Detail & Related papers  (2026-02-12T03:02:22Z)
• Cracking IoT Security: Can LLMs Outsmart Static Analysis Tools? [1.8549313085249322]
  This work presents the first comprehensive evaluation of Large Language Models (LLMs) across a multi-category interaction threat taxonomy.<n>We benchmark Llama 3.1 8B, Llama 70B, GPT-4o, Gemini-2.5-Pro, and DeepSeek-R1 across zero-, one-, and two-shot settings.<n>Our findings show that while LLMs exhibit promising semantic understanding, their accuracy degrades significantly for threats requiring cross-rule structural reasoning.
  arXiv  Detail & Related papers  (2026-01-02T04:17:36Z)
• SoK: a Comprehensive Causality Analysis Framework for Large Language Model Security [9.26421208894097]
  Large Language Models (LLMs) exhibit remarkable capabilities but remain vulnerable to adversarial manipulations such as jailbreaking.<n>We introduce a unified causality analysis framework that supports all levels of causal investigation in LLMs.<n>We empirically evaluate the framework on multiple open-weight models and safety-critical benchmarks including jailbreaks.
  arXiv  Detail & Related papers  (2025-12-04T14:25:15Z)
• Context-Aware Hierarchical Learning: A Two-Step Paradigm towards Safer LLMs [38.3239023969819]
  Large Language Models (LLMs) have emerged as powerful tools for diverse applications.<n>We identify and propose a novel class of vulnerabilities, termed Tool-Completion Attack (TCA)<n>We introduce Context-Aware Hierarchical Learning (CAHL) to address these vulnerabilities.
  arXiv  Detail & Related papers  (2025-12-03T12:10:21Z)
• Cannot See the Forest for the Trees: Invoking Heuristics and Biases to Elicit Irrational Choices of LLMs [83.11815479874447]
  We propose a novel jailbreak attack framework, inspired by cognitive decomposition and biases in human cognition.<n>We employ cognitive decomposition to reduce the complexity of malicious prompts and relevance bias to reorganize prompts.<n>We also introduce a ranking-based harmfulness evaluation metric that surpasses the traditional binary success-or-failure paradigm.
  arXiv  Detail & Related papers  (2025-05-03T05:28:11Z)
• MIRAGE: Multimodal Immersive Reasoning and Guided Exploration for Red-Team Jailbreak Attacks [85.3303135160762]
  MIRAGE is a novel framework that exploits narrative-driven context and role immersion to circumvent safety mechanisms in Multimodal Large Language Models.<n>It achieves state-of-the-art performance, improving attack success rates by up to 17.5% over the best baselines.<n>We demonstrate that role immersion and structured semantic reconstruction can activate inherent model biases, facilitating the model's spontaneous violation of ethical safeguards.
  arXiv  Detail & Related papers  (2025-03-24T20:38:42Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.