Paper Summerizer

This project contains 3 collections of papers that fetch from the arxiv API, namely Machine Learning, Deep Learning, and Reinforcement Learning.

The website is made with Next.js, and the API is with Flask.

1. To build the images for the api and the website, run

   make build
   

2. To run the api and the website, run

   make run
   

3. To terminate the api and the website, run

   make down
   

Future Updates

1. The API will be updated to fetch more papers from the arxiv API.
   • Allow paginations.
2. The API will be rewritten in Golang due to the performance issue.
3. The maintainer is able to a thesis paper to the API, and the paper will be parsed with GROBID.

   GROBID is a machine learning library for extracting, parsing and re-structuring raw documents such as PDF into structured XML/TEI encoded documents with a particular focus on technical and scientific publications.

4. Once the paper is parsed, each section of the paper will be sent to OpenAI to be simplified.
5. The simplification result will be stored in the database.
6. Users can requests for the simplified paper through /api/v1/get endpoint.