Skip to content

beeevita/SPT

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

SPT

Official code for paper Semantic-aware Permutation Training (Mitigating Reversal Curse in Large Language Models via Semantic-aware Permutation Training)

Guidelines

We mainly opensource our scripts for seperating the text into semantic chunks, including the query templates used, a seperation sample (see in seperate.sh), etc. Besides, we put the processed data in the directory data and the raw data in raw_data. If you are interested in the length distribution of the chunks, run seperate/stat.py.

For training framework, we mainly refer to Stanford Alpaca codebase.

TODO

  • collect training datasets used in our experiments (celebrity relations, person description, QA)
  • open seperate code for seperation

About

Official code for paper Semantic-aware Permutation Training

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors