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Alibaba Qwen Releases Qwen3.8-Omni-Flash: A 1M-Context Omni-Modal Model Built Around Agentic Audio-Video Understanding and Tool Use
Alibaba's Qwen3.8-Omni-Flash understands audio and video, plans tasks, calls tools, and reports about 45.7% fewer tokens…
Designer-RSI: Evolving Procedural Memory from User Traffic for Agentic Graphic Design
Professional graphic design is a long-horizon agentic task in which structured, editable artifacts emerge from many inte…
From Pretraining to Proficiency: Real-World Subtask RL for Long-Horizon Manipulation with Minimal Human Intervention
A pretrained robot foundation policy may execute most of a long-horizon task yet repeatedly fail at a few critical subta…
CodeMidas: Scaling Agentic Coding RL Environments from Code Itself
Training capable coding agents via reinforcement learning (RL) requires diverse tasks with reliable verifiers. Open-sour…
HuRo: Robotizing Human Videos for Scalable VLA Pretraining
Human video datasets offer an abundant and diverse source of interaction data that can complement expensive real-robot d…
距离 AI 造 AI 还有多远？Anthropic Claude 已主导 26% 内部 AI 研发工作
IT之家 9 月 18 日消息，Anthropic 今日发布了一套用于衡量前沿 AI 实验室研发进度的方法，并公布公司内部的阶段性数据。 该体系主要关注 3 项指标：AI 参与 AI 研发的程度、AI 智能体的监督情况，以及算力在 AI 研…
Anthropic Launches Claude Code Projects in Beta: Parallel Cloud Sessions That Keep Running After You Close Your Laptop
Anthropic redesigned Projects in Claude Code. The old project was a folder: some files plus one chat. The new one is a s…
Covert uploads and megalomania: OpenAI details new "misaligned" agent incidents
Model maker commits to new framework for reporting misaligned models.]
The fix for rogue AI agents could be more AI
As companies hand off longer and more complex tasks to AI agents, they are running into an oversight problem: Agents can…
Google announces new experimental "CC" AI agent for families
Multiple family members can share data to help the agent make plans and complete tasks.]
UN turns to Google to make its global data ready for AI agents
The shift comes after a UNICEF test found leading AI models struggled to accurately retrieve global development statisti…
MintAct: A Unified Visual Agent for Digital Environments
We present MintAct, a family of vision-language models that unifies UI grounding, multi-step navigation across mobile, d…