Skip to content

Reference Implementation for WSDM 2018 Paper "Hyperbolic Representation Learning for Fast and Efficient Neural Question Answering"

Notifications You must be signed in to change notification settings

svetli-n/WSDM2018_HyperQA

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

48 Commits
 
 
 
 
 
 

Repository files navigation

HyperQA

Code for WSDM 2018 Paper

"Hyperbolic Representation Learning for Fast and Efficient Neural Question Answering" - Yi Tay, Anh Tuan Luu, Siu Cheung Hui - Proceedings of WSDM 2018.

https://arxiv.org/pdf/1707.07847

This repository contains a reference implementation of HyperQA (which is merely copied from the main experiment repository I have). It is helpful for some details that I had no space to report in the actual paper.

I had to strip away some components from other models and also copy paste from my own library. Therefore, some dependencies may be missing for now. A running version with Preprocessing or training scripts will be uploaded when I have time.

Coming Soon. Cleanin in Progress..

Reference

Please cite the WSDM version when it is out in the proceedings.

@article{DBLP:journals/corr/TayLH17a,
  author    = {Yi Tay and
               Anh Tuan Luu and
               Siu Cheung Hui},
  title     = {Enabling Efficient Question Answer Retrieval via Hyperbolic Neural
               Networks},
  journal   = {CoRR},
  volume    = {abs/1707.07847},
  year      = {2017},
  url       = {http://arxiv.org/abs/1707.07847},
  archivePrefix = {arXiv},
  eprint    = {1707.07847},
  timestamp = {Sat, 05 Aug 2017 14:56:20 +0200},
  biburl    = {http://dblp.org/rec/bib/journals/corr/TayLH17a},
  bibsource = {dblp computer science bibliography, http://dblp.org}
}

About

Reference Implementation for WSDM 2018 Paper "Hyperbolic Representation Learning for Fast and Efficient Neural Question Answering"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%