Spaces:
Sleeping
Sleeping
metadata
title: Paper Hero
emoji: 💪
colorFrom: indigo
colorTo: yellow
sdk: docker
app_port: 7860
pinned: true
license: apache-2.0
💪 Paper Hero
A toolkit to help search for papers from aclanthology, arXiv and dblp.
🎉 We have a live demo at hugginface spaces. Check it out [here] !
🌴 Setup
- Make sure you have Git and Python 3.10.8 installed (or Python >= 3.9).
- Install dependencies:
pip install -r requirements.txt,python -m spacy download en_core_web_sm
🚀 QuickStart
Run the example in run.py:
# clone this repo
$ git clone https://github.com/Spico197/paper-hero.git
$ cd paper-hero
# download and install dependencies
$ pip install -r requirements.txt
$ python -m spacy download en_core_web_sm
# get ready for the acl data, since it is cache-based
$ bash scripts/get_aclanthology.sh
$ python run.py
# the results will be saved into `results/`, check them out 🎉
$ ls results
from src.interfaces.aclanthology import AclanthologyPaperList
from src.utils import dump_paper_list_to_markdown_checklist
if __name__ == "__main__":
# use `bash scripts/get_aclanthology.sh` to download and prepare anthology data first
paper_list = AclanthologyPaperList("cache/aclanthology.json")
ee_query = {
"title": [
# Any of the strings below is matched
["information extraction"],
["event", "extraction"], # title must include `event` and `extraction`
["event", "argument", "extraction"],
["event", "detection"],
["event", "classification"],
["event", "tracking"],
["event", "relation", "extraction"],
],
# Besides the title constraint, venue must also meet the needs
"venue": [
["acl"],
["emnlp"],
["naacl"],
["coling"],
["findings"],
["tacl"],
["cl"],
],
}
ee_papers = paper_list.search(ee_query)
dump_paper_list_to_markdown_checklist(ee_papers, "results/ee-paper-list.md")
🪧 Live Demo Usage
🗺️ Roadmap
- aclanthology
- arXiv
- dblp
- add frontend support for building a demo
- year and month searching