Summary of New Emerged Security and Privacy Of Pre-trained Model: a Survey and Outlook, by Meng Yang et al.
New Emerged Security and Privacy of Pre-trained Model: a Survey and Outlook
by Meng Yang, Tianqing Zhu, Chi Liu, WanLei Zhou, Shui Yu, Philip S. Yu
First submitted to arxiv on: 12 Nov 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: None
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 In this paper, researchers explore the unique security challenges posed by pre-trained language processing and computer vision models. Despite their impressive performance, these models can leak sensitive information or generate harmful responses, eroding user trust. To address this gap, the authors propose a taxonomy of attacks and defenses for pre-trained models based on input, weights, and security test scenarios. The taxonomy categorizes methods into No-Change, Input-Change, and Model-Change approaches. This study provides a comprehensive review of existing security issues, highlighting strengths and limitations, and identifies new research opportunities in the security and privacy of pre-trained models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper looks at the risks that come with using super powerful language processing and computer vision models. These models can be very good at doing things like understanding speech or recognizing pictures, but they can also accidentally share private information or say something mean. The people who make these models are trying to figure out how to keep them safe from bad guys who might try to use them for evil. To help with that, the authors made a list of different types of attacks and defenses that people could use to protect the models. |
