LeDat98 / NexusRAG

NexusRAG Hybrid Knowledge Base with Agentic Chat, Citations & Knowledge Graph **Upload documents. Ask questions. Get cited answers.** NexusRAG combines vector search, knowledge graph, and cross-encoder reranking into one seamless RAG pipeline — powered by Gemini, local Ollama, or fully offline sentence-transformers. Features · Quick Start · Model Recommendations · Tech Stack --- Architecture Showcase --- Beyond Traditional RAG Most RAG systems follow a simple pipeline: split text → embed → retrieve → generate. NexusRAG goes further at every stage: | Aspect | Traditional RAG | NexusRAG | |---|---|---| | **Document Parsing** | Plain text extraction, structure lost | Docling: preserves headings, page boundaries, formulas, layout | | **Images & Tables** | Ignored entirely | Extracted, captioned by vision LLM, embedded as searchable vectors | | **Chunking** | Fixed-size splits, breaks mid-sentence | Hybrid semantic + structural chunking (respects headings, tables) | | **Embeddings** | Single model for everything | Dual-model: BAAI/bge-m3 (1024d, search) + KG embedding (Gemini 3072d / Ollama / sentence-transformers) | | **Retrieval** | Vector similarity only | 3-way parallel: Vector over-fetch + KG entity lookup + Cross-encoder rerank | | **Knowledge** | No entity awareness | LightRAG graph: entity extraction, relationship mapping, multi-hop traversal | | **Context** | Raw chunks dumped to LLM | Structured assembly: KG insights → cited chunks → related images/tables | | **Citations** | None or manual | Auto-generated 4-char IDs with page number and heading path | | **Page awareness** | Lost after chunking | Preserved end-to-end: chunk → citation → document viewer navigation | --- Features Deep Document Parsing (Docling) NexusRAG uses Docling for structural document understanding — not just text extraction: • **Structural preservation** — Heading hierarchy ( ), page boundaries, paragraph grouping • **Formula enrichment** — LaTeX math notation preserved during conversion • **Multi-format** — PDF, DOCX, PPTX, HTML, TXT with consistent output • **Hybrid chunking** — respects semantic AND structural boundaries — never splits mid-heading or mid-table • **Page-aware metadata** — Every chunk carries its page number, heading path, and references to images/tables on the same page Hybrid Retrieval Pipeline | Stage | Technology | Details | |---|---|---| | **Vector Embedding** | BAAI/bge-m3 | 1024-dim multilingual bi-encoder (100+ languages) | | **KG Embedding** | Gemini / Ollama / sentence-transformers | Configurable: Gemini (3072d), Ollama, or local sentence-transformers (e.g. bge-m3 1024d) | | **Vector Search** | ChromaDB | Cosine similarity, over-fetch top-20 candidates | | **Knowledge Graph** | LightRAG | Entity/relationship extraction, keyword-to-entity matching | | **Reranking** | BAAI/bge-reranker-v2-m3 | Cross-encoder joint scoring — encodes (query, chunk) pairs together | | **Generation** | Gemini / Ollama | Agentic streaming chat with function calling | **Why two embedding models?** Vector search needs speed (local bge-m3, 1024-dim). Knowledge graph extraction needs semantic richness for entity recognition — choose Gemini Embedding (3072-dim, cloud), Ollama, or sentence-transformers (fully local, no API needed). Each model is optimized for its role. **Retrieval flow:** • **Parallel retrieval** — Vector over-fetch (top-20) + KG entity lookup run simultaneously • **Cross-encoder reranking** — All 20 candidates scored jointly with the query through a transformer (far more precise than cosine similarity alone) • **Filtering** — Keep top-8 above relevance threshold (0.15), with fallback to top-3 if all below • **Media discovery** — Find images and tables on the same pages as retrieved chunks Visual Document Intelligence Images and tables are **embedded into chunk vectors** — not stored separately. When Docling extracts an image on page 5, its LLM-generated caption is appended to the text chunks on that page before embedding. This means searching for "revenue chart" finds chunks that contain the chart description, without needing a separate image search index. **Image Pipeline** • Docling extracts images from PDF/DOCX/PPTX (up to 50 per document, 2x resolution) • Vision LLM (Gemini Vision or Ollama multimodal) generates captions: specific numbers, labels, trends • Captions appended to page chunks: • Chunk is embedded → **image becomes vector-searchable** through its description • During retrieval, images on matched pages are surfaced as references **Table Pipeline** • Docling exports tables as structured Markdown (preserving rows, columns, dimensions) • Text LLM summarizes each table: purpose, key columns, notable values (max 500 chars) • Summaries appended to page chunks: • Table summaries injected back into document Markdown as blockquotes for the document viewer Citation System Every answer is grounded in source documents with **4-character citation IDs** (e.g., ): • **Inline citations** — Clickable badges embedded directly in the answer text • **Source cards** — Each citation shows filename, page number, heading path, and relevance score • **Cross-navigation** — Click a citation to jump to the exact section in the document viewer • **Image references** — Visual content cited separately as with page tracking • **Strict grounding** — The LLM is instructed to only cite sources that directly support claims, max 3 per sentence Knowledge Graph Visualization Interactive force-directed graph built from extracted entities and relationships: • **Entity types** — Person, Organization, Product, Location, Event, Technology, Financial Metric, Date, Regulation (configurable) • **Force simulation** — Repulsion + spring forces + center gravity with real-time physics • **Pan & zoom** — Mouse drag, scroll wheel (0.3x-3x), keyboard reset • **Node interaction** — Click to select, hover to highlight connected edges, drag to reposition • **Entity scaling** — Node radius proportional to co…