RAG_Techniques and HiRAG
These are **complements**: NirDiamant/RAG_Techniques provides a broad survey of RAG implementation patterns and methodologies, while HiRAG represents a specific advanced technique (hierarchical knowledge retrieval) that could be studied as one instantiation or integrated with the techniques in the survey repository.
About RAG_Techniques
NirDiamant/RAG_Techniques
This repository showcases various advanced techniques for Retrieval-Augmented Generation (RAG) systems. RAG systems combine information retrieval with generative models to provide accurate and contextually rich responses.
Covers advanced RAG patterns including agentic retrieval loops, hybrid search strategies (dense-sparse retrieval fusion), query optimization techniques, and multi-document reasoning—beyond basic retrieval pipelines. Implementations target popular frameworks like LangChain and LlamaIndex with code-first Jupyter notebooks, focusing on practical enhancements for production-grade systems.
About HiRAG
hhy-huang/HiRAG
[EMNLP'25 findings] This is the official repo for the paper, HiRAG: Retrieval-Augmented Generation with Hierarchical Knowledge.
Organizes retrieved documents into local, global, and bridge knowledge layers within a graph database, enabling multi-level contextual retrieval that significantly outperforms flat naive RAG and graph-based approaches across comprehensiveness, empowerment, and diversity metrics. Supports LLM-agnostic integration through pluggable APIs (DeepSeek, ChatGLM, OpenAI) and batch async embedding with configurable caching, while providing modular retrieval modes (hierarchical, local-only, global-only, bridge-only) for controlled knowledge access. Includes evaluation pipelines on the UltraDomain benchmark across specialized domains (Mix, CS, Legal, Agriculture) with LLM-based assessment scoring.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work