~ similar to 2606.23370· 20 results
Sina Abdollahi, Mohammad M Maheri, Javad Forough, Amir Al Sadi +4 more
AgenTEE is a system that enables the secure, confidential execution of complex LLM agent pipelines directly on edge devices by using isolated confidential virtual machines.
Guanlong Wu, Zhaohan li, Yao Zhang, Zheng Zhang +3 more
CachePrune introduces a privacy-aware, fine-grained KV cache sharing mechanism that allows LLM inference systems to safely reuse cache entries across users' requests, significantly improving efficienc…
CloakLM is a software-only memory-obfuscation framework that protects large language models from model exfiltration attacks.
Chanwoo Cho, Wooseok Kim, Yonglak Son, Young Seo Lee +1 more
The paper proposes Voltron, a framework for executing large language model inferences on multiple user-end devices at the edge to achieve higher accuracy and satisfy QoS requirements.
Guanyu Cai, Ruiming Tian, Lang Yang, Zhouhong Ren +3 more
This paper presents a comprehensive measurement study on the performance and energy consumption of large language models on mobile devices, using five frameworks and three hardware backends, and intro…
The paper introduces POE2, an extension of POE for intra-process isolation using memory protection keys and new registers, providing stronger security guarantees but introducing architectural complexi…
Di Lu, Qingwen Zhang, Yujia Liu, Xuewen Dong +3 more
The paper introduces EBCC, an OCI-compatible runtime architecture that manages composite confidential-computing workloads by integrating TEE-backed execution into the standard container lifecycle.
This paper introduces execution-state capsules for low-latency, small-batch, on-device AI serving.
Yunze Zhao, Yibo Zhao, Yuchen Zhang, Zaoxing Liu +1 more
The paper introduces GRIEF, a greybox fuzzer that discovers critical, concurrency-related vulnerabilities in LLM serving systems by treating timed multi-request traces as inputs, finding issues like c…
Zhe Liu, Zonghao Ying, Wenxin Zhang, Quanchen Zou +4 more
SafeHarbor is a novel, hierarchical memory-augmented framework that establishes context-aware decision boundaries for LLM agents, achieving state-of-the-art safety while minimizing over-refusal.
LiteAtt introduces a verifier-less, Peer-to-Peer Self-Attestation (P2P-SA) framework for modern IoT MCUs, enabling mutual authentication and firmware attestation directly within the connection handsha…
Katharine Daly, Yu Xiao, Zachary Garrett, Brett McLarnon +8 more
The paper presents a next-generation Federated Learning system that provides externally verifiable central Differential Privacy guarantees while improving device coverage and privacy-utility tradeoff.
This paper provides the first comprehensive review of threats and defenses specifically targeting on-device AI inference, revealing a significant imbalance where certain attack types, like adversarial…
MemLineage introduces a novel, cryptographically-backed defense mechanism that enforces a chain-of-custody for LLM agent memory, preventing untrusted or poisoned state from justifying sensitive action…
This paper replicates and extends a study on Java security API misuse in LLMs, finding that while newer models improve performance, the misuse risk persists and is significantly mitigated by external…
This survey analyzes the unique security threats posed by complex, multi-agent AI systems and proposes Confidential Computing (CC) using Trusted Execution Environments (TEEs) as a hardware-rooted defe…
The paper proposes a two-stage post-training pipeline to create a small, local LLM agent (PrivEsc-LLM) capable of performing Linux privilege escalation, achieving high success rates while drastically…
TrustANN provides query-level trust for distributed approximate nearest-neighbor retrieval without modifying the underlying algorithm.
The paper presents a secure and open Cloud Inference Service (CIS) framework called OpenPCC that uses commercially available Trusted Execution Environments (TEEs) to protect user data.