sentence-transformers-query-processing

An AI-powered research assistant that can understand documents, retrieve relevant context, and answer questions using Retrieval-Augmented Generation (RAG).

An intelligent Agentic RAG chatbot that goes beyond basic document Q&A with adaptive retrieval, multi-document search, conversation memory, and query refinement for more relevant answers. Includes source citations, execution tracing, RAG evaluation, security guardrails, and authentication. Built with Streamlit, LangGraph, Chroma, Groq, and FastAPI

RAG(Knowledge assistant) with source citations, hybrid BM25 + dense search, cross-encoder reranking, and retrieval metrics (Recall@K, MRR, nDCG).

Neurosymbolic semantic query answering demo using domain glossaries, SBERT embeddings, and soft homomorphism matching.