Image de couverture de livre

StructBreak: Structural Cognitive Overload-Induced Safety Failures in MLLMs

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Luo, Yang, Liu, Xinran, Ji, Tiantian, Yin, Zhiyi, Peng, Lingyun, Li, Shuyu
Format:	Preprint
Publié:	2026
Sujets:	Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

ShadowMerge: A Novel Poisoning Attack on Graph-Based Agent Memory via Relation-Channel Conflicts
par: Luo, Yang, et autres
Publié: (2026)

Breaking Bad Molecules: Are MLLMs Ready for Structure-Level Molecular Detoxification?
par: Lin, Fei, et autres
Publié: (2025)

GuardAD: Safeguarding Autonomous Driving MLLMs via Markovian Safety Logic
par: Zhang, Tianyuan, et autres
Publié: (2026)

StructVRM: Aligning Multimodal Reasoning with Structured and Verifiable Reward Models
par: Zhang, Xiangxiang, et autres
Publié: (2025)

CSR-Bench: A Benchmark for Evaluating the Cross-modal Safety and Reliability of MLLMs
par: Liu, Yuxuan, et autres
Publié: (2026)

On-Policy Self-Evolution via Failure Trajectories for Agentic Safety Alignment
par: Yin, Bo, et autres
Publié: (2026)

Mitigating Societal Cognitive Overload in the Age of AI: Challenges and Directions
par: Lahlou, Salem
Publié: (2025)

The Forgotten Shield: Safety Grafting in Parameter-Space for Medical MLLMs
par: Zhao, Jiale, et autres
Publié: (2025)

StructEval: Benchmarking LLMs' Capabilities to Generate Structural Outputs
par: Yang, Jialin, et autres
Publié: (2025)

Probing Implicit Bias in Semi-gradient Q-learning: Visualizing the Effective Loss Landscapes via the Fokker--Planck Equation
par: Yin, Shuyu, et autres
Publié: (2024)

Struct-X: Enhancing Large Language Models Reasoning with Structured Data
par: Tan, Xiaoyu, et autres
Publié: (2024)

StructSynth: Leveraging LLMs for Structure-Aware Tabular Data Synthesis in Low-Data Regimes
par: Liu, Siyi, et autres
Publié: (2025)

OmniStruct: Universal Text-to-Structure Generation across Diverse Schemas
par: Huang, James Y., et autres
Publié: (2025)

Explaining and Breaking the Safety-Helpfulness Ceiling via Preference Dimensional Expansion
par: Huang, ShiYing, et autres
Publié: (2026)

Pragma-VL: Towards a Pragmatic Arbitration of Safety and Helpfulness in MLLMs
par: Wen, Ming, et autres
Publié: (2026)

VeriStruct: AI-assisted Automated Verification of Data-Structure Modules in Verus
par: Sun, Chuyue, et autres
Publié: (2025)

AESOP: Adversarial Execution-path Selection to Overload Deep Learning Pipelines
par: Li, Tingxi, et autres
Publié: (2026)

StreamingBench: Assessing the Gap for MLLMs to Achieve Streaming Video Understanding
par: Lin, Junming, et autres
Publié: (2024)

FairReason: Balancing Reasoning and Social Bias in MLLMs
par: Pan, Zhenyu, et autres
Publié: (2025)

StructRAG: Boosting Knowledge Intensive Reasoning of LLMs via Inference-time Hybrid Information Structurization
par: Li, Zhuoqun, et autres
Publié: (2024)

A priori Estimates for Deep Residual Network in Continuous-time Reinforcement Learning
par: Yin, Shuyu, et autres
Publié: (2024)

Low-Resource Safety Failures Are Action Failures, Not Representation Failures
par: Aziz, Rashad, et autres
Publié: (2026)

FlexStructRAG: Flexible Structure-Aware Multi-Granular Relational Retrieval for RAG
par: Chen, Mengzhu, et autres
Publié: (2026)

StructXLIP: Enhancing Vision-language Models with Multimodal Structural Cues
par: Ruan, Zanxi, et autres
Publié: (2026)

StructLens: A Structural Lens for Language Models via Maximum Spanning Trees
par: Sakajo, Haruki, et autres
Publié: (2026)

Image2Struct: Benchmarking Structure Extraction for Vision-Language Models
par: Roberts, Josselin Somerville, et autres
Publié: (2024)

RL-Struct: A Lightweight Reinforcement Learning Framework for Reliable Structured Output in LLMs
par: Hu, Ruike, et autres
Publié: (2025)

Enhancing Cognitive Diagnosis by Modeling Learner Cognitive Structure State
par: Chen, Zhifu, et autres
Publié: (2024)

Diffusion-CAM: Faithful Visual Explanations for dMLLMs
par: Zuo, Haomin, et autres
Publié: (2026)

StructRL: Recovering Dynamic Programming Structure from Learning Dynamics in Distributional Reinforcement Learning
par: Nowak, Ivo
Publié: (2026)

StructPrune: Structured Global Pruning asymptotics with $\mathcal{O}(\sqrt{N})$ GPU Memory
par: Song, Xinyuan, et autres
Publié: (2025)

OmniVideoBench: Towards Audio-Visual Understanding Evaluation for Omni MLLMs
par: Li, Caorui, et autres
Publié: (2025)

The Geometry of Alignment Collapse: When Fine-Tuning Breaks Safety
par: Springer, Max, et autres
Publié: (2026)

When Context Flips, Safety Breaks: Diagnosing Brittle Safety in Aligned Language Models
par: Choi, Dasol, et autres
Publié: (2026)

StructEval: Deepen and Broaden Large Language Model Assessment via Structured Evaluation
par: Cao, Boxi, et autres
Publié: (2024)

StructMem: Structured Memory for Long-Horizon Behavior in LLMs
par: Xu, Buqiang, et autres
Publié: (2026)

Affordance Benchmark for MLLMs
par: Wang, Junying, et autres
Publié: (2025)

Redundancy Principles for MLLMs Benchmarks
par: Zhang, Zicheng, et autres
Publié: (2025)

Breaking the Martingale Curse: Multi-Agent Debate via Asymmetric Cognitive Potential Energy
par: Liu, Yuhan, et autres
Publié: (2026)

From Specific-MLLMs to Omni-MLLMs: A Survey on MLLMs Aligned with Multi-modalities
par: Jiang, Shixin, et autres
Publié: (2024)