This repository is a curated list of academic papers focusing on machine learning security and privacy. It targets researchers and practitioners in the field, providing a structured overview of cutting-edge research published in top-tier security conferences. The benefit is a centralized, categorized resource for staying updated on ML security threats and defenses.

How It Works

The repository organizes papers by sub-topics within ML security and privacy, such as adversarial attacks, data poisoning, backdoor attacks, and various privacy concerns like membership inference and model extraction. Each entry includes the paper title, conference, year, relevant keywords, and direct links to the PDF and code repositories where available. This structured approach facilitates efficient discovery and access to relevant research.

Quick Start & Requirements

This is a curated list, not a software package. No installation or execution is required. Access is via web browser to the GitHub repository.

Highlighted Details

• Comprehensive categorization covering 2 major areas (Security, Privacy) and dozens of sub-topics.
• Links to both PDF and code for a significant portion of the listed papers.
• Focus on papers from top-tier security conferences (IEEE S&P, ACM CCS, USENIX Security, NDSS).
• Includes a dedicated section for Large Language Model (LLM) security and privacy.

Maintenance & Community

The list is primarily maintained by Ping He from NESA Lab. Contributions are welcomed.

Licensing & Compatibility

Copyright is held by gnipping, with all rights reserved. This repository is for informational purposes and does not impose restrictions on commercial use or closed-source linking beyond standard copyright law.

Limitations & Caveats

The repository is a static list and does not provide any tools or frameworks for ML security or privacy. Its value is solely in its curation and organization of existing research.