Inter-agent communication is essential to multi-agent language-model systems, yet a single message may combine task-critical information with instructions not authorized by the original request. Prompt-based defenses leave enforcement to models exposed to adversarial messages, while indiscriminate message removal disca...
Jinghan Xu, Longze Fan, Zeyuan Wang et al.· 0 citations
These results give the widely repeated qualitative recommendation of "balanced human-AI collaboration" a precise, testable form and suggest an interior-optimum capacity ratio as a concrete design target for security operations centers (SOCs), including those securing IT/OT-converged critical infrastructure.
Mustafa S. Aljumaily, Hayder Kareem Abed, Nawar S. Alseelawi· 0 citations
AI-generated imagery evolves faster than benchmark-specific detector evaluations, making a single score an incomplete account of generalization. This paper evaluates Neural Defend ARCAS 1B across benchmark families without benchmark-specific parameter updates. We retain native aggregation and supplement it with record-...
Sivashankar Selvarajan, Piyush Verma, Sumit Kumar et al.· 0 citations
Organizations fine-tune small language models on private data and then compress them to 4 bits for resource-efficient deployment. We show that the compression method also affects privacy. What separates the methods is not the bit width but whether they tune their rounding on a small sample of text, the calibration corp...
Cristhian Kapelinski, Diego Kreutz· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Structured multi-agent workflows exchange intermediate messages whose content and form can reveal private state even when the final output is safe. We identify selection-channel leakage: after authorization fixes what may be released, a private-state-aware choice among semantically valid realizations creates an additio...
Jing-Heng Xu, Long-Ze Fan, Ze-Yuan Wang et al.· 0 citations
This work introduces KEX-bench, a benchmark for evaluating coding agents on exploit primitive generation against real operating-system kernels, and evaluates state-of-the-art coding agents paired with frontier and open-weight models under fixed tool-call budgets.
Junyoung Jang, Gwanhyun Lee, Hwiwon Lee et al.· 0 citations
A conditional decision-preservation proposition: successful local admission implies that a specified synchronous policy would authorize the same action at the admission point, provided approval is sound, all policy dependencies are represented and current, observations are faithful, and consumption is atomic.
This work proposes skill habit formation, a model that mines its own execution history for candidate skills, deterministic variants that compete against the incumbent rather than replacing it, and measures what this costs in accuracy.
To the authors' knowledge, this is the first end-to-end autonomous attack-remediation demonstration on bare-metal industrial devices, establishing controlled feasibility - not zero-day discovery or production-OT transfer.
Adel Elzemity, Budi Arief, Shujun Li et al.· arXiv.org· 1 citation
BridgeShield is presented, a graph-based framework for detecting cross-chain bridge attacks through risk-aware modeling of cross-chain execution behaviors, and consistently outperforms existing rule-based and graph-based baselines in cross-chain attack detection.
Dan-yan Lin, Shun-Feng Lu, Zi-Yan Liu et al.· 0 citations
Large language models (LLMs) are increasingly deployed in production systems, raising concerns about their exposure to adversarial manipulation through prompt injection and jailbreak attacks. Classifier-based guardrails, such as Prompt Guard 2, are widely used as a first line of defense against such attacks, but their...
Fernando Outeda, Gustavo Betarte, J. Campo et al.· 0 citations
Large Language Models (LLMs) are increasingly used in cybersecurity, where accurate analysis often requires multi-step and context-dependent reasoning over complex and heterogeneous data. However, existing prompting approaches typically focus on eliciting reasoning without explicitly considering how intermediate reason...
Ji-Ling Zhou, Aisvarya Adeseye, Antti Hakkala et al.· 0 citations