README.md

paper.py

A command-line-based arXiv.org article browser and batch downloader.

$ paper.py --publish-years 2015-2019 'ti:"attention is all you need"' | fold -w 80
1
-
ID: 1706.03762v7
Authors: ['Ashish Vaswani', 'Noam Shazeer', 'Niki Parmar', 'Jakob Uszkoreit', 'L
lion Jones', 'Aidan N. Gomez', 'Lukasz Kaiser', 'Illia Polosukhin']
Title: Attention Is All You Need
Published: Mon Jun 12 2017 05:57:34PM UTC
Updated: Wed Aug 02 2023 12:41:18AM UTC
Output file name: 2017-Ashish_Vaswani-Noam_Shazeer-Niki_Parmar-Jakob_Uszkoreit-L
lion_Jones-Aidan_N_Gomez-Lukasz_Kaiser-Illia_Polosukhin-Attention_Is_All_You_Nee
d-1706.03762v7.pdf
PDF links:
['http://arxiv.org/pdf/1706.03762v7']
Other links:
['http://arxiv.org/abs/1706.03762v7']

2
-
ID: 1910.14537v3
Authors: ['Sufeng Duan', 'Hai Zhao']
Title: Attention Is All You Need for Chinese Word Segmentation
Published: Thu Oct 31 2019 03:32:19PM UTC
Updated: Tue Oct 06 2020 06:38:42AM UTC
Output file name: 2019-Sufeng_Duan-Hai_Zhao-Attention_Is_All_You_Need_for_Chines
e_Word_Segmentation-1910.14537v3.pdf
PDF links:
['http://arxiv.org/pdf/1910.14537v3']
Other links:
['http://arxiv.org/abs/1910.14537v3']

3
-
ID: 1906.02792v1
Authors: ['Manjot Bilkhu', 'Siyang Wang', 'Tushar Dobhal']
Title: Attention is all you need for Videos: Self-attention based Video
  Summarization using Universal Transformers
Published: Thu Jun 06 2019 07:59:56PM UTC
Updated: Thu Jun 06 2019 07:59:56PM UTC
Output file name: 2019-Manjot_Bilkhu-Siyang_Wang-Tushar_Dobhal-Attention_is_all_
you_need_for_Videos_Self_attention_based_Video_Summarization_using_Universal_Tra
nsformers-1906.02792v1.pdf
PDF links:
['http://arxiv.org/pdf/1906.02792v1']
Other links:
['http://arxiv.org/abs/1906.02792v1']

$ paper.py --download --download-selection 1 \
> --publish-years 2015-2019 'ti:"attention is all you need"' >/dev/null
$ ls | fold -w 80
2017-Ashish_Vaswani-Noam_Shazeer-Niki_Parmar-Jakob_Uszkoreit-Llion_Jones-Aidan_N
_Gomez-Lukasz_Kaiser-Illia_Polosukhin-Attention_Is_All_You_Need-1706.03762v7.pdf

Setup

1. Install Python3 (tested using v3.11.2).
2. Clone this repository.
3. Copy the file paper.py to some directory that is in the PATH environment variable of your Operating system.
4. Optional: add the following convenience functions and aliases to the configuration of your POSIX-shell-compatible OS shell:

   arxivh()
   {
       cat <<EOF
   arxivall <query> [option...]
   arxivtitle <query> [option...]
   arxivauthor <query> [option...]
   arxivabstract <query> [option...]
   arxivcomment <query> [option...]
   arxivjournalreference <query> [option...]
   arxivsubjectcategory <query> [option...]
   arxivreportnumber <query> [option...]
   arxivid <query> [option...]
   
   For arxivXYZ, the <query> parameter specifies a query of type XYZ.
   
   [option...] is a list of options to be passed to paper.py.
   
   The query for arxivauthor should be an author name in the following form:
   <lowercase-last-name>_<first-name-initial>_<optional-middle-name-initial>
   The query for author "Art I. Ficial" should be "ficial_a_i".
   
   If the last name comprises of multiple words, then they must be separated by underscores.
   For example, the query for "Fictional Del Maestro" should be "del_maestro_f".
   EOF
   }
   _arxiv_parent()
   {
       ARXIV_QUERY_TYPE="$1"
       ARXIV_QUERY="$2"
       shift 2
   
       paper.py "$@" "${ARXIV_QUERY_TYPE}:${ARXIV_QUERY}"
   }
   alias arxivall='_arxiv_parent all'
   alias arxivtitle='_arxiv_parent ti'
   alias arxivauthor='_arxiv_parent au'
   alias arxivabstract='_arxiv_parent abs'
   alias arxivcomment='_arxiv_parent co'
   alias arxivjournalreference='_arxiv_parent jr'
   alias arxivsubjectcategory='_arxiv_parent cat'
   alias arxivreportnumber='_arxiv_parent rn'
   # alias arxivid='_arxiv_parent id'
   arxivid()
   {
       paper.py --id-list "$@"
   }

Help

Run paper.py -h to get a list of fine-tuning options, usage examples, and relevant sources for arXiv.org search query construction.

Basic usage examples

Search for articles by author Art I. Ficial:

$ paper.py "au:ficial_a_i"

The list looks good? Download the articles (pdf):

$ paper.py -d "au:ficial_a_i"

Look up articles by their arXiv ids...

$ paper.py -i 2309.06314,1811.02452

... and download them:

$ paper.py -d -i 2309.06314,1811.02452

Search for articles by title:

$ paper.py "ti:quantum"

Combine them:

$ paper.py "au:ficial_a_i AND ti:quantum"
$ paper.py "au:ficial_a_i AND (ti:quantum OR ti:subconvexity)"
$ paper.py "au:ficial_a_i ANDNOT (ti:quantum OR ti:subconvexity)"
$ paper.py -i 2309.06314,1811.02452 'au:ficial_a_i AND ti:"quantum unique ergodicity"'

Acknowledgment

Inspired by GNU Emacs-based ideas of @ultronozm.