Skip to content

ACL2024 Integrating Multi-scale Contextualized Information for Byte-based Neural Machine Translation

Notifications You must be signed in to change notification settings

ictnlp/Multiscale-Contextualization

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Multiscale-Contextualization

Description

Codes and scripts for ACL2024 Findings paper "Integrating Multi-scale Contextualized Information for Byte-based Neural Machine Translation".

Changes to fairseq files

  • add the file multiscale_transformer.py to fairseq/models
  • add the file multiscale_attention.py to fairseq/modules
  • add the file transformer_multiscale_layer.py to fairseq/modules
  • change fairseq/modules/init.py to import the added packages

How to use it?

Add pathes to the scripts in the scripts folder, and run them.

About

ACL2024 Integrating Multi-scale Contextualized Information for Byte-based Neural Machine Translation

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published