File size: 2,096 Bytes
09c527b |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 |
---
license: llama3.1
datasets:
- BAAI/Infinity-Instruct
base_model:
- meta-llama/Llama-3.1-8B-Instruct
tags:
- Instruct_Tuning
library_name: transformers
pipeline_tag: text-generation
---
# Shadow-FT
<a href="https://arxiv.org/pdf/2505.12716"><b>[📜 Paper]</b></a> •
<a href="https://huggingface.co/collections/taki555/shadow-ft-683288b49e1e5e1edcf03135"><b>[🤗 HF Models]</b></a> •
<a href="https://github.com/wutaiqiang/Shadow-FT"><b>[🐱 GitHub]</b></a>
This repo contains the weights from our paper: <a href="https://arxiv.org/pdf/2505.12716" target="_blank">Shadow-FT: Tuning Instruct via Base</a> by <a href="https://wutaiqiang.github.io" target="_blank">Taiqiang Wu*</a> <a href="https://rummyyang.github.io/" target="_blank">Runming Yang*</a>, Jiayi Li, Pengfei Hu, Ngai Wong and Yujiu Yang.
\* for equal contributions.
## Overview
<img src="framework.png" width="100%" />
Observation:
- Directly tuning the INSTRUCT (i.e., instruction tuned) models often leads to marginal improvements and even performance degeneration.
- Paired BASE models, the foundation for these INSTRUCT variants, contain highly similar weight values (i.e., less than 2% on average for Llama 3.1 8B).
$\Rightarrow$ We propose the Shadow-FT framework to tune the INSTRUCT models by leveraging the corresponding BASE models. The key insight is to fine-tune the BASE model, and then _directly_ graft the learned weight updates to the INSTRUCT model.
## Performance
This repository contains the Llama-3.1-8B tuned on BAAI-2k subsets using Shadow-FT.
<img src="performance.png" width="100%" />
please refer to [our paper](https://arxiv.org/pdf/2505.12716) for details.
## ☕️ Citation
If you find this repository helpful, please consider citing our paper:
```
@article{wu2025shadow,
title={Shadow-FT: Tuning Instruct via Base},
author={Wu, Taiqiang and Yang, Runming and Li, Jiayi and Hu, Pengfei and Wong, Ngai and Yang, Yujiu},
journal={arXiv preprint arXiv:2505.12716},
year={2025}
}
```
For any questions, please pull an issue or email at `[email protected]`
|