Skip to content

NYCU-MLLab/LCD

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Learnable Contrastive Decoding in State-Space LLMs

Environment Build

We recommmend using uv to build the environment.

uv sync

check uv installation

Code Usage

main.ipynb is the training scripts. We train the LCD module from RWKV.\ test.ipynb is for testing the result over 3 models: RWKV, Transformer (here we use pythia-70m) and our approach LCD

Datasets

The used training/testing dataset are provided in link.

About

Learnable Contrastive Decoding in State-Space LLMs

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •