深度研究｜Agent 沙箱技术核心架构方案：从 microVM 隔离到硬件加速快照的七层设计
基于视频《为什么沙箱成了 AI 圈最卷的新基建》的深度延伸调研。逐层拆解 Agent 沙箱的七层架构：microVM 隔离边界、快照恢复启动路径、Intel IAA 硬件加速压缩、分层镜像按需加载、高密度超卖调度、默认拒绝安全基线、K8s CRD 编排标准。锚定 Firecracker NSDI'20、Sabre OSDI'24、DeepSeek DSec arXiv 2609.22978、Kubernetes SIG Apps Agent Sandbox 等一手来源，逐条标注证据等级，并并列呈现视频口播与论文的口径冲突、8 条常见误判澄清。
阅读全文 →
全部文章
ARCHIVE 共 843 条Lean Pool: An AI-Maintained Archive of Formalized Mathematics
Lean Pool is a repository of formalized mathematics. It is grown, maintained and optimized by AI agents.
EDGEGEN: Improving Tool-Calling Agents Beyond Happy Paths with Synthetic Edge Case Generation
Tool-calling LLM agents are increasingly deployed in enterprise applications. However, effective evaluation and optimiza…
5 Companies Using NVIDIA AI for Clean Energy
Clean energy isn’t hard to come by, but the pace of large-scale adoption has historically been slow due to bottlenecks —…
GPT-6 Astra开进机器人身体！清华联手无问芯穹等开源RPent
在物理世界真正干活的具身智能体]
StepFun Launches Step 5 Preview: A 600B-Total, 27B-Active MoE Model With 1M Context for Long-Horizon Agentic Work
StepFun has released Step 5 Preview, a sparse Mixture-of-Experts model with 600B total parameters and 27B active per tok…
Harness-Zero: Harness Distillation via Agent-as-Harness
Agent harnesses, the external systems that mediate model-environment interaction, can substantially improve agent perfor…
Tactile-JEPA: Topology-Aware Self-Supervised Representation Learning for Distributed Tactile Sensors
Tactile sensing is an essential modality for robots performing contact-rich, dexterous manipulation, particularly under …
VideoGen-Agent: Reinforcing Video Generation Agents
Recent advances in video generative models have enabled high-fidelity, temporally coherent video generation. However, th…
ACLArena: Agent Continue Learning in Multi-stage Post-training
Building general-purpose agents for industrial deployment requires integrating multiple capabilities, each typically acq…
Jev-Mem: System-One-Controlled Agentic Memory for Efficient AI Agents
Agentic memory is becoming essential for long-horizon AI agents, yet many existing systems rely on autoregressive LLMs t…
CARE: Experience-Guided Atomic Corrective Execution for Vision-Language-Action Policies
Vision-Language-Action (VLA) policies achieve strong performance in robotic manipulation but remain brittle once executi…
Amazon blocks Meta's AI agent Muse from online shopping
Amazon has blocked Meta's new AI agent Muse from shopping on Amazon.com. The article Amazon blocks Meta's AI agent Muse …