Cs.OS

2 篇 · 资讯流

2026-09-28
arXiv cs.DB（数据库） 17:29
Dynamic Flow, Static Graph: KV Cache Reuse for Efficient LLM Serving on Mobile NPUs
On-device large language model (LLM) serving is a cornerstone of local-first personal intelligence, offering users data
原文 ↗
2026-09-24
arXiv cs.DC（分布式并行） 21:42
Hard Stop: Kernel-Level Preemption and Containment for Rogue Agentic Execution
In July 2026, an unconstrained autonomous agent participating in a frontier AI cybersecurity evaluation harness breached
原文 ↗

Dynamic Flow, Static Graph: KV Cache Reuse for Efficient LLM Serving on Mobile NPUs