Summary

This project offers an intelligent academic paper writing system designed to automate the generation of scholarly content. It addresses the challenge of drafting academic papers by leveraging multiple large language models (LLMs) to produce chapter drafts from input proposals or research designs, serving as a valuable reference or starting point for researchers and students. The system aims to streamline the writing process, ensure logical coherence, and maintain academic formatting standards.

How It Works

The system employs a multi-model collaboration strategy, integrating APIs from providers like OpenAI, Anthropic (Claude), and Alibaba (Tongyi Qianwen). It utilizes a multi-stage processing pipeline, incorporating network search capabilities via Serper or Tavily APIs for academic material retrieval. Core functionalities include automated generation of paper sections based on input documents, quality control mechanisms, and outputting content in standard academic Word document format using python-docx. This approach aims to enhance efficiency and consistency in academic writing.

Quick Start & Requirements

• Installation: Clone the repository, create and activate a Python 3.8+ virtual environment, then run pip install -r requirements.txt.
• Prerequisites: Requires API keys for OpenAI and Claude (mandatory), and optionally for Alibaba, Serper, and Tavily. Input documents must be in .docx format and placed in the input/ directory.
• Usage: Prepare a .docx proposal in input/, modify pipeline.py with the input path, and execute python pipeline.py. Generated papers are saved in output/.