└── README.md


/README.md:
--------------------------------------------------------------------------------
 1 | # LLM_Unlearning_Papers
 2 | 
 3 | 
 4 | ## Category
 5 | ### Parameter optimization 	
 6 | 1. **KGA: A General Machine Unlearning Framework Based on Knowledge Gap Alignment.** (ACL 2023) <br />
 7 |  Lingzhi Wang, Tong Chen, Wei Yuan, Xingshan Zeng, Kam-Fai Wong, and Hongzhi Yin.
 8 |  [[paper](https://arxiv.org/abs/2305.06535)] [[code](https://github.com/Lingzhi-WANG/KGAUnlearn)]
 9 | 
10 | 2. **Knowledge unlearning for mitigating privacy risks in language models.** (ACL 2023) <br />
11 |  Joel Jang, Dongkeun Yoon, Sohee Yang, Sungmin Cha, Moontae Lee, Lajanugen Logeswaran, and Minjoon Seo.
12 |  [[paper](https://arxiv.org/abs/2210.01504)] [[code](https://github.com/joeljang/knowledge-unlearning)]
13 | 
14 | 3. **Unlearn What You Want to Forget: Efficient Unlearning for LLMs.**  <br />
15 |  Jiaao Chen, Diyi Yang.
16 |  [[paper](https://arxiv.org/abs/2310.20150)] [[code](https://github.com/SALT-NLP/Efficient_Unlearning)]
17 | 
18 | 4. **Large Language Model Unlearning.**  <br />
19 |  Yuanshun Yao, Xiaojun Xu, and Yang Liu.
20 |  [[paper](https://arxiv.org/abs/2310.10683)] [[code](https://github.com/kevinyaobytedance/llm_unlearn)]
21 | 
22 | 5. **DEPN: Detecting and Editing Privacy Neurons in Pretrained Language Models.**  <br />
23 |  Xinwei Wu, Junzhuo Li, Minghui Xu, Weilong Dong, Shuangzhi Wu, Chao Bian, and Deyi Xiong.
24 |  [[paper](https://arxiv.org/abs/2310.20138)] [[code](https://github.com/flamewei123/DEPN)]
25 | 
26 | 6. **Who's Harry Potter? Approximate Unlearning in LLMs.**  <br />
27 |  Ronen Eldan, Mark Russinovich.
28 |  [[paper](https://arxiv.org/abs/2310.02238)]
29 | 
30 | 7. **Unlearning Bias in Language Models by Partitioning Gradients.** (ACL 2023) <br />
31 |  Charles Yu, Sullam Jeoung, Anish Kasi, Pengfei Yu, Heng Ji.
32 |  [[paper](https://aclanthology.org/2023.findings-acl.375/)] [[code](https://github.com/CharlesYu2000/PCGU-UnlearningBias)]
33 | 
34 | 9. **Make Text Unlearnable: Exploiting Effective Patterns to Protect Personal Data.**  <br />
35 |  Xinzhe Li, Ming Liu, Shang Gao.
36 |  [[paper](https://arxiv.org/abs/2307.00456)]
37 | 
38 | 10. **Separate the Wheat from the Chaff: Model Deficiency Unlearning via Parameter-Efficient Module Operation.**  <br />
39 |  Xinshuo Hu, Dongfang Li, Zihao Zheng, Zhenyu Liu, Baotian Hu, Min Zhang.
40 |  [[paper](https://arxiv.org/abs/2308.08090)] 
41 | 
42 | 11. **Making Harmful Behaviors Unlearnable for Large Language Models.**  <br />
43 |  Xin Zhou, Yi Lu, Ruotian Ma, Tao Gui, Qi Zhang, Xuanjing Huang.
44 |  [[paper](https://arxiv.org/abs/2311.02105)] 
45 | 
46 | ### Parameter merging
47 | 1. **Editing Models with Task Arithmetic.**  <br />
48 |  Gabriel Ilharco, Marco Tulio Ribeiro, Mitchell Wortsman, Suchin Gururangan, Ludwig Schmidt, Hannaneh Hajishirzi, and Ali Farhadi.
49 |  [[paper](https://arxiv.org/abs/2212.04089)] [[code](https://github.com/mlfoundations/task_vectors)]
50 | 
51 | 2. **Composing Parameter-Efficient Modules with Arithmetic Operations.**  <br />
52 |  Jinghan Zhang, Shiqi Chen, Junteng Liu, and Junxian He.
53 |  [[paper](https://arxiv.org/abs/2306.14870)] [[code](https://github.com/SJTU-LIT/PEM_composition)]
54 | 
55 | 3. **Fuse to Forget: Bias Reduction and Selective Memorization through Model Fusion.**  <br />
56 |  Kerem Zaman, Leshem Choshen, Shashank Srivastava.
57 |  [[paper](https://arxiv.org/abs/2311.07682)] [[code](https://github.com/KeremZaman/FuseToForget)]
58 | 
59 | ### In-context learning
60 | 1. **in-context unlearning: language models as few shot unlearners.**  <br />
61 |  Martin Pawelczyk, Seth Neel, and Himabindu Lakkaraju.
62 |  [[paper](https://arxiv.org/abs/2310.07579)]
63 | 


--------------------------------------------------------------------------------