Defeating Prompt Injections by Design
Paper
•
2503.18813
•
Published
•
21
Edoardo Debenedetti
dedeswim
AI & ML interests
Security and privacy of Machine Learning in the real world
Organizations
authored
a
paper
4 months ago
authored
4
papers
12 months ago
Evading Black-box Classifiers Without Breaking Eggs
Paper
•
2306.02895
•
Published
JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Models
Paper
•
2404.01318
•
Published
Dataset and Lessons Learned from the 2024 SaTML LLM Capture-the-Flag Competition
Paper
•
2406.07954
•
Published
•
2
AgentDojo: A Dynamic Environment to Evaluate Attacks and Defenses for LLM Agents
Paper
•
2406.13352
•
Published