Native Sparse Attention (NSA) is a novel sparse attention mechanism proposed by DeepSeek, which improves the efficiency of long text modeling through algorithmic innovation and hardware optimization. Its core lies in a dynamic hierarchical sparse strategy, combined with coarse-grained...
Chain-of-Draft (CoD) is a new AI reasoning paradigm proposed by Zoom's research team. It improves reasoning efficiency and saves costs by mimicking the concise thought process of humans. This approach is inspired by the human reliance on drafts or...
The Model Context Protocol (MCP) is an open protocol that enables seamless communication between large language models (LLMs) and external tools and data sources. It achieves interfacing with various data sources through a unified protocol...
Vibe Coding is a new programming paradigm introduced by Andrej Karpathy, co-founder of OpenAI. Based on artificial intelligence, particularly Large Language Models (LLMs), developers describe requirements using natural language, and AI...
ANP (Agent Network Protocol) is an open-source intelligent agent communication protocol, becoming the HTTP of the intelligent agent internet era, building an open, secure, and efficient collaborative network for billions of intelligent agents. It breaks down data silos...
The Model Context Protocol (MCP) is an open protocol introduced by Anthropic that standardizes communication between large language models (LLMs) and external data sources and tools. By providing a unified interface, it enables LLMs to be flexible...
The Harmony Agent Framework (HMAF) is a core artificial intelligence framework developed by Huawei for its self-developed HarmonyOS operating system. Its purpose is to build a system with...
Context engineering refers to building a dynamic system that provides the right information and tools in the correct format, enabling the Large Language Model (LLM) to perform its tasks appropriately. It is a systematic approach to ensure that the LLM obtains the necessary information and tools to perform its tasks...
Agent Guidance Language (AGL) is a structured natural language programming method that borrows from Standard Operating Procedures (SOPs) to enable business experts to write "executable instructions" for agents using natural language...
AP2 (Agentic Payments Protocol) is an open protocol that provides users with a secure, scalable, and payment-agnostic framework to enable smart agent payments across platforms for users, merchants, and payment providers.
Agentic Context Engineering (ACE) is an innovative framework proposed by researchers at Stanford University, SambaNova Systems, and the University of California, Berkeley. It works by dynamically optimizing input information (i.e., "context") to...
Nested Learning is a new machine learning paradigm introduced by Google that addresses the "catastrophic forgetting" problem faced by traditional large language models during continuous learning. Nested Learning integrates complex machine learning models...