Cover Image

The Capability Paradox: How Smarter Auditors Make Multi-Agent Systems Less Secure

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Liu, Qiqi, Holz, Thorsten, Ye, Shilin, Song, Runhan
Format:	Preprint
Published:	2026
Subjects:	Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

AgentAuditor: Human-Level Safety and Security Evaluation for LLM Agents
by: Luo, Hanjun, et al.
Published: (2025)

Multi-Agent Security Tax: Trading Off Security and Collaboration Capabilities in Multi-Agent Systems
by: Peigne-Lefebvre, Pierre, et al.
Published: (2025)

ClawLess: A Security Model of AI Agents
by: Lu, Hongyi, et al.
Published: (2026)

The Optimization Paradox in Clinical AI Multi-Agent Systems
by: Bedi, Suhana, et al.
Published: (2025)

The Security Cost of Intelligence: AI Capability, Cyber Risk, and Deployment Paradox
by: Choi, Sukwoong
Published: (2026)

No More, No Less: Task Alignment in Terminal Agents
by: Mavali, Sina, et al.
Published: (2026)

HexaCoder: Secure Code Generation via Oracle-Guided Synthetic Training Data
by: Hajipour, Hossein, et al.
Published: (2024)

Debug Smarter, Not Harder: AI Agents for Error Resolution in Computational Notebooks
by: Grotov, Konstantin, et al.
Published: (2024)

Diffusion Models for Smarter UAVs: Decision-Making and Modeling
by: Emami, Yousef, et al.
Published: (2025)

Select Smarter, Not More: Prompt-Aware Evaluation Scheduling with Submodular Guarantees
by: Ma, Xiaoyu, et al.
Published: (2026)

DarkForest: Less Talk, Higher Accuracy for Multi-Agent LLMs
by: Li, Yi, et al.
Published: (2026)

Personal LLM Agents: Insights and Survey about the Capability, Efficiency and Security
by: Li, Yuanchun, et al.
Published: (2024)

Sentinel Agents for Secure and Trustworthy Agentic AI in Multi-Agent Systems
by: Gosmar, Diego, et al.
Published: (2025)

CapSeal: Capability-Sealed Secret Mediation for Secure Agent Execution
by: Jin, Shutong, et al.
Published: (2026)

Tree of Agents: Improving Long-Context Capabilities of Large Language Models through Multi-Perspective Reasoning
by: Yu, Song, et al.
Published: (2025)

Open Challenges in Multi-Agent Security: Towards Secure Systems of Interacting AI Agents
by: de Witt, Christian Schroeder, et al.
Published: (2025)

Ask, Clarify, Optimize: Human-LLM Agent Collaboration for Smarter Inventory Control
by: Duan, Yaqi, et al.
Published: (2025)

Multi-Objective Bayesian Optimization for Networked Black-Box Systems: A Path to Greener Profits and Smarter Designs
by: Kudva, Akshay, et al.
Published: (2025)

ExploitGym: Can AI Agents Turn Security Vulnerabilities into Real Attacks?
by: Wang, Zhun, et al.
Published: (2026)

CADMAS-CTX: Contextual Capability Calibration for Multi-Agent Delegation
by: Qiao, Chuhan
Published: (2026)

Multi-Agent Deep Research: Training Multi-Agent Systems with M-GRPO
by: Hong, Haoyang, et al.
Published: (2025)

Hybrid Agentic AI and Multi-Agent Systems in Smart Manufacturing
by: Farahani, Mojtaba A., et al.
Published: (2025)

MindWatcher: Toward Smarter Multimodal Tool-Integrated Reasoning
by: Chen, Jiawei, et al.
Published: (2025)

Towards Unifying Quantitative Security Benchmarking for Multi Agent Systems
by: Sharma, Gauri, et al.
Published: (2025)

CUAAudit: Meta-Evaluation of Vision-Language Models as Auditors of Autonomous Computer-Use Agents
by: Sumyk, Marta, et al.
Published: (2026)

Political Bias Audits of LLMs Capture Sycophancy to the Inferred Auditor
by: Törnberg, Petter, et al.
Published: (2026)

OxyGent: Making Multi-Agent Systems Modular, Observable, and Evolvable via Oxy Abstraction
by: Hu, Junxing, et al.
Published: (2026)

High Accuracy, Less Talk (HALT): Reliable LLMs through Capability-Aligned Finetuning
by: Franzmeyer, Tim, et al.
Published: (2025)

Think Smarter not Harder: Adaptive Reasoning with Inference Aware Optimization
by: Yu, Zishun, et al.
Published: (2025)

Do Self-Evolving Agents Forget? Capability Degradation and Preservation in Lifelong LLM Agent Adaptation
by: Yu, Ye, et al.
Published: (2026)

SUGAR: Leveraging Contextual Confidence for Smarter Retrieval
by: Zubkova, Hanna, et al.
Published: (2025)

Parallelized Planning-Acting for Efficient LLM-based Multi-Agent Systems in Minecraft
by: Li, Yaoru, et al.
Published: (2025)

How Social is It? A Benchmark for LLMs' Capabilities in Multi-user Multi-turn Social Agent Tasks
by: Wu, Yusen, et al.
Published: (2025)

TransferTOD: A Generalizable Chinese Multi-Domain Task-Oriented Dialogue System with Transfer Capabilities
by: Zhang, Ming, et al.
Published: (2024)

MAFE: Enabling Equitable Algorithm Design in Multi-Agent Multi-Stage Decision-Making Systems
by: Lazri, Zachary McBride, et al.
Published: (2025)

Causal Explanations for Sequential Decision-Making in Multi-Agent Systems
by: Gyevnar, Balint, et al.
Published: (2023)

LIMO: Less is More for Reasoning
by: Ye, Yixin, et al.
Published: (2025)

Foundation-Model-Based Agents in Industrial Automation: Purposes, Capabilities, and Open Challenges
by: Henkel, Vincent, et al.
Published: (2026)

Can Learned Optimization Make Reinforcement Learning Less Difficult?
by: Goldie, Alexander David, et al.
Published: (2024)

Determinants of LLM-assisted Decision-Making
by: Eigner, Eva, et al.
Published: (2024)