Collective-Knowledge-Base
SpaceX Collective Knowledge Base: company-wide engineering notes, failure write-ups and hard-won lessons, structured for Grokbot ingestion.
- 181
- TypeScript
- Apache-2.0
- 2 d ago
SpaceX Collective Knowledge Base: company-wide engineering notes, failure write-ups and hard-won lessons, structured for Grokbot ingestion.
AWS-native knowledge graph RAG framework. Runs GraphRAG, LightRAG, HippoRAG 2 and RAPTOR on one shared index, each selectable at index and query time, with hybrid BM25 + vector + graph retrieval and source provenance. Amazon Bedrock, Neptune and OpenSearch, or fully local. Incremental indexing, multilingual, bring-your-own KG, built-in evaluation.
Multimodal RAPTOR for Disaster Documents using ColVBERT & BLIP. Hierarchical retrieval system over 46 tsunami-related PDFs (2378 pages), combining BLIP-based image captioning, ColVBERT embeddings, and GPT-OSS-20b long-context summarization. Optimized for fast multimodal tree construction and disaster knowledge preservation.
Looped hybrid RAG lab about the cost of a second look: local models, visible traces, published metrics.
Treg免疫細胞系譜を例に、RAPTORアルゴリズムを実装したGPU加速対応のRAG(Retrieval-Augmented Generation)システムです。実際に、5ノードから14ノードを実現しました。
Cancer-RAPTOR : GPU-accelerated hierarchical search system for cancer medical information
🌳 Open-source RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval - Complete open-source implementation with 100% local LLMs (Granite Code 8B + mxbai-embed-large)
PubMedから実際のTreg関連文献を収集し、Enhanced Treg Vocabulary(7層316用語)を用いて高精度に分類。品質評価指標(Silhouette + DBI)に基づく最適クラスタリングでRAPTORツリーを構築します。
Visual RAPTOR ColBERT Integration System - Multimodal document retrieval with SigLIP, PyMuPDF, and evaluation metrics.
An advanced RAG (Retrieval-Augmented Generation) system using RAPTOR algorithm to hierarchically organize and retrieve lessons from the 2011 Great East Japan Earthquake and Tsunami for educational purposes.