Spaces:
Running
Running
Update README.md
Browse files
README.md
CHANGED
@@ -7,4 +7,33 @@ sdk: static
|
|
7 |
pinned: false
|
8 |
---
|
9 |
|
10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
7 |
pinned: false
|
8 |
---
|
9 |
|
10 |
+
# π§ OpenUnlearning Hub: A Collection of Trained/Unlearned LLMs
|
11 |
+
|
12 |
+
Welcome to the **OpenUnlearning Hub**, a central repository of models trained and unlearned using the [OpenUnlearning](https://github.com/locuslab/open-unlearning) framework β a standardized toolkit for benchmarking and accelerating machine unlearning in large language models (LLMs).
|
13 |
+
---
|
14 |
+
|
15 |
+
## π What is OpenUnlearning?
|
16 |
+
|
17 |
+
**OpenUnlearning** is a unified and extensible framework for:
|
18 |
+
- Evaluating unlearning methods and metrics
|
19 |
+
- Comparing faithfulness and efficiency of forgetting algorithms
|
20 |
+
- Providing a common benchmark to accelerate research in LLM unlearning
|
21 |
+
|
22 |
+
Read our paper for the full details:
|
23 |
+
π [arXiv:2506.12618](https://arxiv.org/abs/2506.12618)
|
24 |
+
|
25 |
+
---
|
26 |
+
|
27 |
+
## π£ Citation
|
28 |
+
|
29 |
+
If you use our models or code in your research or applications, please cite:
|
30 |
+
|
31 |
+
```bibtex
|
32 |
+
@article{openunlearning2025,
|
33 |
+
title={{OpenUnlearning}: Accelerating {LLM} Unlearning via Unified Benchmarking of Methods and Metrics},
|
34 |
+
author={Dorna, Vineeth and Mekala, Anmol and Zhao, Wenlong and McCallum, Andrew and Lipton, Zachary C and Kolter, J Zico and Maini, Pratyush},
|
35 |
+
journal={arXiv preprint arXiv:2506.12618},
|
36 |
+
year={2025},
|
37 |
+
}
|
38 |
+
|
39 |
+
|