Modern AI agent harnesses expose lifecycle hooks that bind shell commands to runtime events such as session start, tool calls, and file edits. These commands run with host privileges yet ship as lifecycle-hook configuration and may fire at times the LLM never observes. We identify the lifecycle-hook update path, which...
Pengxun Li, Litian Zhang, Jianwei Hou et al.· 0 citations
This paper introduces geometric indistinguishability as a conceptual lens for a regime in which privacy-induced dispersion in client updates can make minority-class signals harder for robust aggregation to preserve and suggests that aggregation-aware modeling and sample-aware evaluation are promising directions for tru...
Adrita Rahman Tory, Abm Shawkat Ali, M. Layek et al.· 0 citations
Topology-guided safeguards for LLM-based multi-agent systems (MAS) train a GNN over the inter-agent communication graph to localize risky agents and intervene on the topology---but they assume one operator can pool all labeled traces. Across organizations that assumption breaks: episodes contain private prompts, tool o...
Jin-Xi Yu, Eric Jiang, Levina Li et al.· 0 citations
This paper proposes GEO Defender, a two-stage defense aligned with the attack chain that requires no fine-tuning of the target LLM's source use at inference, and demonstrates that GEO Defender reduces the average attack success rate, retains 94.12% of benign-evidence use, preserves answer quality, and generalizes to un...
Hao-Zhang Li, Yang-Guang Shao, Xin-Jie Lin et al.· 1 citation
Reach audiences
Advertise in front of researchers, engineers, and readers.
This work introduces Privatehub, which uses contrastive learning within a diffusion model to generate synthetic multi-sensor streams that keep non-private applications detectable while concealing private ones in sensor-intensive environments.
Jie-Chao Gao, Yuan-Dong Pan, Jie Wang et al.· 0 citations
This work proposes a federated inference framework in which several commercial LLM APIs collaborate without requiring access to raw student data or proprietary model internals, and conducts rigorous privacy-utility analysis showing strong privacy guarantees with minimal accuracy loss.
A privacy firewall pipeline based on a U-Net encoder-decoder, trained entirely on synthetic data, that removes speech from ambient audio while preserving environmental sounds indicative of daily activities is presented.
This paper systematizes LLM watermarking as provenance infrastructure for large-scale data ecosystems and introduces a Big Data Watermarking Readiness framework centered on four deployment workloads: online generation, streaming detection, transformation pipelines, and ecosystem governance.
Huy Phan, Kieu Dang, Ojaswi Dulal et al.· 0 citations
A defense benchmark that pairs 247 human-verified, quality-gated queries with information-preserving and information-distorting GEO rewrites, and evaluates defenses on attack success rate (ASR), false positive rate, and answer quality across three victim LLMs is presented.
This work designs a system that downsamples point cloud content at the origin server and applies partial encryption at the client, and decrypted and upscaled using an ML-based super-resolution model, which effectively reconstructs the original full-resolution point clouds with minimal error and modest inference time.
Mohammad Waquas Usmani, Sankalpa Timilsina, Michael Zink et al.· IEEE International Symposium...· 2 citations
This work introduces CodePoisonRAG, a targeted upstream knowledge-poisoning framework that transforms benign fixed-code entries into poisoned artifacts and shows that RACG poisoning extends beyond the incidental propagation of existing vulnerabilities to the targeted construction and propagation of attacker-selected we...
Varun Gadey, Ziad Marey, Alexandra Dmitrienko· 1 citation
SPADE --- the SPaT Attack Detection and Evaluation dataset --- a labelled, multi-modal, simulation-based dataset designed specifically for deep learning IDS research in this space to support reproducible and comparative IDS research in C-V2X security.
James Di Novo, Hany Ragab, Sylvain P Leblanc· 0 citations
