arxiv paper downloader

env

pip install -r requirements.txt

Windows system meets pdf2image related errors, see issue_markdown for solution.

usage

Download Arxiv papers

Download the latest n day research paper PDF, generate corresponding markdown and preview images.

python arxiv_spider.py --category cs.CL --root_dir  /your/papers/dir/path  --days n

If you want to overwrite previous markdowns, use

python arxiv_spider.py --category cs.CL --root_dir  /your/papers/dir/path  --days n  --overwrite

If your disk space is not enogh and you just want to have a first preview of the main idea of the paper, you can just keep the 8 first pages of the paper. Just use --keep_eight_pages to keep the first eight pages.

python arxiv_spider.py --category cs.CL --root_dir  /your/papers/dir/path  --days n  --keep_eight_pages

todo

• Highlight target conferences and authors.
• Highlight custom keyword list.
• Use VPN Arg Parse.
• Customize the specific areas of interest, such as CS.CL and CS.CV.
• Maintain an index database of all papers for quick search and reference.
• Demo videos.

Useful Arxiv Related Repos

• https://github.com/lukasschwab/arxiv.py
• https://github.com/braun-steven/arxiv-downloader

Download ACL papers

This repo also support acl anthology papaer download.

Step 1: download all papers, default is ACL 2023 papers, you can specify your acl url.

python acl_spider.py

Step 2: filter papers according to your keywords, and get the abstract of every paper.

python filter_papers.py --keywords keyword1 keyword2 keyword3