Matthew Jagielski

机构：Anthropic (United States) · ORCID：0000-0002-9749-0696

发表论文 93 篇 · 总被引 2899 次 · h-index 21

代表论文

• PaLM 2 Technical Report (2023 · arXiv (Cornell University) · 被引 168)
• Poisoning Web-Scale Training Datasets is Practical (2024 · 被引 119)
• Scalable Extraction of Training Data from (Production) Language Models (2023 · arXiv (Cornell University) · 被引 83)
• Advancing Differential Privacy: Where We Are Now and Future Directions for Real-World Deployment (2024 · Harvard Data Science Review · 被引 40)
• Are aligned neural networks adversarially aligned? (2023 · arXiv (Cornell University) · 被引 25)
• SNAP: Efficient Extraction of Private Properties with Poisoning (2023 · 被引 16)