人工智能
83 篇 · 资讯流阶跃新旗舰 Step 5 Preview 发布，跻身 AA 榜单全球开源前三
9 月 20 日，阶跃星辰发布新一代旗舰基座模型 Step 5 Preview，重点面向 AI 编程、软件工程、专业知识工作、金融等场景，提升模型在真实世界 Agentic 任务中的综合表现，并进一步拓展智能与成本之间的最优边界。在最新的全
原文 ↗General Collaborative Intelligence: Architecting Cognition for Resilient Multi-Agent Ecosystems
Multi-agent unmanned systems are moving from isolated, ego-centric sensing toward collaborative intelligence, in which d
原文 ↗Beyond Single-Model Injection: A Threat Model and Defense Architecture for Prompt Injection in Multi-Agent Systems
Existing prompt injection research focuses on single-model chatbot scenarios, where an attacker manipulates one LLM thro
原文 ↗The Weight Is Over - Interactive Diffusion on Consumer GPUs
On-device inference is booming, but the momentum is almost all in language models. Diffusion pipelines are memory hungry
原文 ↗别再让 Agent 裸连数据库了：如何用 MCP + 语义网关解决 NL2SQL 幻觉与权限失控
在 2026 年，给 AI Agent（Cursor、Claude Desktop、Dify、Coze 等）外挂一个数据库连接，已经成为很多开发者的标配操作。从简单的官方 SQLite/MySQL MCP 到各种通用客户端扩展，让大模型“直
原文 ↗Micro-Collaborative Poisoning: A Distributed Attack on RAG Systems
Retrieval-Augmented Generation (RAG) improves large language models by grounding outputs in external knowledge sources,
原文 ↗华为云 All in Agentic Cloud，灵衢昇腾 950 智算集群全球上线
华为云这届全联接大会的调子，是"Agentic Cloud"。华为云 CEO 周跃峰演讲里那句被反复引用的判断是：智能体时代基础设施的核心，不再只是给算力，而是"让每一个 Token 更高效"。 最能说明态度的莫过于那个出圈的名字——灵衢昇
原文 ↗Omni Demand Understanding: A Benchmark for Contextual User-Intent Inference in Multimodal Interaction
Natural audio-visual interaction is emerging as an important interface for AI assistants, allowing users to communicate
原文 ↗Large Language Models as Falsifiers for Cyber-Physical Systems
Falsification searches for counterexamples to formal specifications in cyber-physical systems (CPS). With specifications
原文 ↗Uni-LaDiR: Latent Diffusion Unifies Multimodal Reasoning
Multimodal reasoning requires models to draw on information from multiple modalities throughout the reasoning process. Y
原文 ↗