Skip to content

Luc4Gui/FSPAD

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

FSPAD

Official Pytorch Implementation of Boosting Lossless Speculative Decoding via Feature Sampling and Partial Alignment Distillation

Citation

You are encouraged to modify/distribute this code. However, please acknowledge this code and cite the paper appropriately.

@article{gui2024boosting,
  title={Boosting Lossless Speculative Decoding via Feature Sampling and Partial Alignment Distillation},
  author={Gui, Lujun and Xiao, Bin and Su, Lei and Chen, Weipeng},
  journal={arXiv preprint arXiv:2408.15562},
  year={2024}
}

Acknowledgments

This code uses libraries from EAGLE and Medusa repository.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published