ChatMarx  by TommyZihao

University political education Q&A AI

created 2 years ago
251 stars

Top 99.8% on SourcePulse

GitHubView on GitHub
Project Summary

This project provides a ChatGPT-like question-answering AI for university ideological and political education courses, specifically covering modern Chinese history, ethics, Mao Zedong Thought, and Marxism. It targets students and educators seeking an AI assistant to query official textbooks, offering five distinct AI personas for tailored responses.

How It Works

The system leverages large language models and Chinese NLP techniques, integrating with the Langchain framework. It processes four specific 2023 edition university textbooks, enabling users to ask questions and receive answers based on the textbook content. The AI personas (Tutor, Scholar, Knowledge Point, Researcher, Lu Xun) offer different interaction styles and functionalities, with some personas focusing on direct textbook retrieval and others on analyzing current events.

Quick Start & Requirements

  • Install: The README does not provide specific installation instructions but links to a demo and Bilibili video.
  • Prerequisites: Requires access to the four specified 2023 edition university ideological and political education textbooks. It utilizes various Chinese NLP models from Hugging Face (e.g., text2vec-base-chinese, m3e-base) and LLMs (e.g., InternLM, ChatGLM2-6B).
  • Demo: An online demo is available at https://chatmarx.zihaoai.cn.

Highlighted Details

  • Features five distinct AI personas: Tutor, Scholar, Knowledge Point, Researcher, and Lu Xun.
  • Tutor persona provides direct quotes and sources from the textbooks.
  • Researcher and Lu Xun personas analyze and comment on current social events.
  • Developed by a team with expertise in AI, deep learning, and software engineering, including a Microsoft MVP.

Maintenance & Community

The project is led by Zhang Zihao (同济子豪兄), a recognized figure in AI education with a significant Bilibili presence. Contact for technical consultation or business cooperation is available via Bilibili private message.

Licensing & Compatibility

The project is explicitly stated to be for non-profit, educational, and research purposes only. No specific open-source license is mentioned, implying potential restrictions on commercial use or integration into closed-source systems.

Limitations & Caveats

The AI-generated answers may require further optimization and do not represent the views of the development team. The project is intended for non-profit use, which may limit its applicability in commercial or closed-source environments.

Health Check
Last commit

2 years ago

Responsiveness

Inactive

Pull Requests (30d)
0
Issues (30d)
0
Star History
1 stars in the last 30 days

Explore Similar Projects

Starred by Aravind Srinivas Aravind Srinivas(Cofounder of Perplexity), Shizhe Diao Shizhe Diao(Research Scientist at NVIDIA; Author of LMFlow), and
12 more.

gpt-3 by openai

0.0%
16k
Research paper on large language model few-shot learning
created 5 years ago
updated 4 years ago
Feedback? Help us improve.