TL;DR: NEAT is a novel PEFT method that uses a lightweight neural network to directly model cumulative weight updates for adaptation targets and capture intricate and non-coherent structures.
The experiment setups are located in each subfolder.
-
Notifications
You must be signed in to change notification settings - Fork 1
yibozhong/neat
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
About
Code for paper [Neat: Nonlinear Parameter-efficient Adaptation of Pre-trained Models]
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published