GithubHelp home page GithubHelp logo

xhsun1997 / cmrc2019 Goto Github PK

View Code? Open in Web Editor NEW

This project forked from ymcui/cmrc2019

0.0 1.0 0.0 10.38 MB

The Third Evaluation Workshop on Chinese Machine Reading Comprehension (CMRC 2019)

Home Page: https://hfl-rc.github.io/cmrc2019/

License: Creative Commons Attribution Share Alike 4.0 International

Python 99.68% Shell 0.32%

cmrc2019's Introduction

The Third Evaluation Workshop on Chinese Machine Reading Comprehension (CMRC 2019)

This is the official repository for CMRC 2019.

More details:

Updates

April 8, 2020 Add dataset description paper
August 1, 2019 Qualifying data has been released, check data directory. (Note that, json key 'answers' no longer contains answers.)
June 10, 2019 Development data has been released, check data directory.
May 24, 2019 Baseline system has been released, check baseline directory.
May 23, 2019 Train/Trial data have been released, check data directory.
May 17, 2019 We are expected to release task introduction and call for participation on late May or early June. Stay tuned!

Directory Guide

  • baseline: a Chinese BERT-based simple baseline system

  • eval: contains official evaluation script

  • data: contains offical evaluation data

  • sample_submission: sample submission for codalab competition platform (trial_rand_submission.zip is a randomly generated prediction file, trial_submission.zip is the BERT baseline prediction file)

Baseline System

We provide a BERT-based baseline system for participants (check baseline directory for more info).

Results on other sets will be annouced later.

QAC: Question-Level Accuracy

PAC: Passage-Level Accuracy

Data Passage # Query # QAC PAC Fake Candidates Availability
Trial Data 139 1,504 71.941% 28.776% No Public
Train Data 9,638 100,009 N/A N/A No Public
Development Data 300 3,053 70.586% 13.333% Yes Public
Qualifying Data 500 5,081 70.01% 8.20% Yes Semi-Hidden
Test Data - - - - Yes Hidden

Reference

If you wish to use our data in your research, please cite our paper:

@article{cui-2020-cmrc2019,
  title={A Sentence Cloze Dataset for Chinese Machine Reading Comprehension},
  author={Cui, Yiming and Liu, Ting and Yang, Ziqing and Chen, Zhipeng and Ma, Wentao and Che, Wanxiang and Wang, Shijin and Hu, Guoping},
  journal={arXiv preprint arXiv:2004.03116},
  year={2020}
}

Organization Committee

Host: Chinese Information Processing Society of China (CIPS)
Organizer: Joint Laboratory of HIT and iFLYTEK (HFL)
Sponsor: iFLYTEK Co., Ltd. and iFLYTEK Research (Hebei)

Evaluation Co-Chairs

Ting Liu, Harbin Institute of Technology
Yiming Cui, iFLYTEK Research

Official Discussion QQ Group

You can join our QQ group for discussion purpose.

qqgroup.png

Official HFL WeChat Account

Follow Joint Laboratory of HIT and iFLYTEK Research (HFL) on WeChat.

qrcode.png

Contact us

Any problems? Feel free to concat us.
Email: cmrc2019 [aT] 126 [DoT] com
Forum: CodaLab Competition Forum

cmrc2019's People

Contributors

ymcui avatar

Watchers

James Cloos avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.