Reimagining service delivery in the agentic era with Google Public Sector
cloud.google.com·21h ago
AI Research
arXiv CS.AI·Jun 10·5 min read
Advancing the State-of-the-Art in Empirical Privacy Auditing
TL;DR
Large language models (LLMs) can inadvertently memorize sensitive training data, posing privacy risks. A new method generates synthetic 'canary' examples to enhance empirical privacy auditing (EPA) by measuring data leakage.
✦ Why It Matters
Engineers can implement synthetic canaries to enhance privacy audits in machine learning models effectively.
Key Takeaways
- Utilize high-temperature sampling to generate synthetic canaries.
- Incorporate canaries into privacy audits for better risk assessment.
- Fine-tune auxiliary models on synthetic data for robust audits.
- Investigate model capacity effects on memorization and privacy.
How It Works
The proposed method generates synthetic canaries by sampling from LLMs at high temperatures, which creates diverse and identifiable examples. These canaries are mixed with sensitive training data to test the model's ability to protect privacy.
Since the canaries are non-private, they can be reused without compromising real data, enhancing the robustness of privacy audits.