搜索结果：Markdown格式

Tonic Validate Logging is a Python library that enables developers to log RAG application outputs to the Tonic Validate platform for automated evaluation using LLM-assisted metrics, visualization, and

RAG-Anything is an open-source framework developed by HKU researchers that enables unified multimodal retrieval-augmented generation, allowing AI systems to understand and retrieve knowledge from text

LLMII is an open-source tool that uses local AI models to automatically generate captions and keywords for images, embedding this metadata directly into image files for enhanced searchability and orga

This research systematically evaluates quantization and dimensionality reduction techniques for RAG vector embeddings, finding that float8 quantization combined with PCA offers optimal 8x compression

GEO (Generative Engine Optimization) is the new paradigm for content visibility in the AI era, shifting focus from keyword ranking for search engines to semantic understanding and authority building f

Satya is an offline-first educational platform that integrates Retrieval-Augmented Generation (RAG) with the Phi 1.5 language model, designed to run locally on standard hardware (4GB RAM) without internet dependency, specifically addressing educational infrastructure gaps in rural areas. 原文翻译： Satya是一个离线优先的教育平台，集成了检索增强生成（RAG）与Phi 1.5语言模型，旨在无需互联网依赖，在标准硬件（4GB RAM）上本地运行，专门解决农村地区的教育基础设施差距。

Memora is a lightweight MCP server that provides AI agents with persistent memory storage, knowledge graph visualization, semantic search, and cross-session context management through SQLite, cloud sync, and vector embeddings. 原文翻译： Memora是一个轻量级MCP服务器，通过SQLite、云同步和向量嵌入技术，为AI代理提供持久记忆存储、知识图谱可视化、语义搜索和跨会话上下文管理。

This article explores an alternative approach to the Model Context Protocol (MCP) for integrating tools with Large Language Models (LLMs) by leveraging existing OpenAPI servers. It proposes a simpler, more intuitive method that uses structured HTTP API definitions as tool inputs, requiring only minimal authentication flow additions. The implementation is demonstrated through a concise Scala script, focusing on core tool integration while omitting MCP's broader features like prompts and resources. 原文翻译： 本文探讨了一种替代模型上下文协议（MCP）的方法，通过利用现有的OpenAPI服务器为大型语言模型（LLM）集成工具。它提出了一种更简单、更直观的方法，使用结构化的HTTP API定义作为工具输入，仅需添加最小的身份验证流程。通过一个简洁的Scala脚本演示了实现，专注于核心工具集成，同时省略了MCP更广泛的功能，如提示和资源。

LLM Knowledge Graph Builder transforms unstructured data into Neo4j knowledge graphs, offering superior accuracy over traditional RAG through GraphRAG's relationship-aware retrieval mechanism. 原文翻译： LLM知识图谱构建器将非结构化数据转换为Neo4j知识图谱，通过GraphRAG的关系感知检索机制，提供比传统RAG更优越的准确性。

This survey provides a comprehensive framework for optimizing LLM agent workflows through agentic computation graphs (ACGs), distinguishing between static and dynamic methods based on when workflow structure is determined, and proposing structure-aware evaluation metrics beyond traditional task performance. 原文翻译： 本综述通过智能体计算图（ACGs）为优化LLM智能体工作流提供了一个全面的框架，根据工作流结构确定的时间区分静态和动态方法，并提出了超越传统任务性能的结构感知评估指标。