DeepTutor

Tracked

A multi-agent personalized learning system integrating RAG, knowledge graphs, and interactive visualizations.

Author HKUDS Open Sourced 2025-12-28 Last Commit Unknown

Overview

DeepTutor is an agent-native, open-source personalized tutoring platform developed by HKUDS. It combines Retrieval-Augmented Generation (RAG), knowledge graphs, and multi-agent collaborative reasoning to deliver end-to-end learning support from knowledge retrieval to practice and assessment.

Key Features

  • Large-scale document Q&A with cited answers powered by vector retrieval and RAG pipelines.
  • Multi-agent problem solving with a dual-loop architecture supporting real-time streaming reasoning.
  • Intelligent exercise generation that produces and validates practice questions by difficulty and exam style.
  • Interactive learning visualization that transforms complex concepts into step-by-step demonstrations.

Use Cases

  • University teaching and online course platforms where instructors build question banks and mock exams.
  • Self-learners who benefit from interactive explanations and personalized practice sessions.
  • Researchers conducting literature reviews and systematic reviews with deep retrieval and report generation.

Technical Details

  • Built with Python/FastAPI backend and Next.js frontend, supporting Docker deployment and local development.
  • Retrieval layer combines embeddings with knowledge graph structures for semantic search.
  • Parallelized dynamic task queue with centralized citation management and plugin-style tool integrations.