Beyond Memorization: Violating Pivacy Via Inference With Large Language Models
PrePrint, arXiv, October 11, 2023. Beyond Memorization: Violating Privacy Via Inference With Large Language Models. Robin Staab, Mark Vero, Mislav Balunovic, Martin Vechev, Department of Computer Science, ETH Zurich. “Current privacy research on large language models (LLMs) primarily focuses on the issue of extracting memorized training data. At the same time, models’ inference capabilities have …