Summary of Forget to Flourish: Leveraging Machine-unlearning on Pretrained Language Models For Privacy Leakage, by Md Rafi Ur Rashid et al.
Forget to Flourish: Leveraging Machine-Unlearning on Pretrained Language Models for Privacy Leakage
by Md Rafi Ur Rashid, Jing Liu, Toshiaki Koike-Akino, Shagufta Mehnaz, Ye Wang
First submitted to arxiv on: 30 Aug 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper introduces a novel poisoning technique that manipulates large language models to increase the leakage of private data during fine-tuning. This approach, called model-unlearning, enhances membership inference and data extraction attacks while preserving model utility. The authors demonstrate the effectiveness of this attack across different models, datasets, and fine-tuning setups, highlighting potential risks for users who download pre-trained models from unverified sources. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This study shows that using pre-trained language models from unknown sources can be risky because they can be intentionally designed to expose sensitive information. A new way to make these models leak private data is discovered, which makes it easier to figure out if someone’s dataset is being used and what’s in it. The authors tested this method on different models and datasets and found that it works well. |
Keywords
» Artificial intelligence » Fine tuning » Inference
