Skip to content

A Reasoning Network for multi-relation Question Answering over Cross-lingual Knowledge Graphs

License

Notifications You must be signed in to change notification settings

tan92hl/Cross-lingual-Reasoning-Network-for-CLKGQA

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Cross-lingual-Reasoning-Network-for-CLKGQA

A Reasoning Network for multi-relation Question Answering over Cross-lingual Knowledge Graphs

CLKGQA

This is the repo hosting all the experiment code of the paper "CLRN: A Reasoning Network for multi-relation Question Answering overCross-lingual Knowledge Graphs"

Data

MLPQ

Environment

  • cudatoolkit=10.2
  • pytorch==1.8.1
  • transformers==4.5.0
  • chinese_converter==1.0.2
  • jsonlines==2.0.0
  • auto_mix_prep==0.2.0
  • protobuf==3.15.8
  • python_Levenshtein==0.12.2

NVIDIA GeForce RTX 2080 GPU is used in our experiments

CLRN

This module is consisted of multilingual-BERT and CLRN and a training pipeline to fine tune BERT and train the CLRN simultaneously. The trained models are saved to the specified directory after the training is finished.

python train.py

question_answering

The question answering module utilizes the trained model from CLRN module and reason continuously across knowledge graphs of different languages and save the generated equal paths and aligned entities as the answering result.

python main.py

About

A Reasoning Network for multi-relation Question Answering over Cross-lingual Knowledge Graphs

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages