Gemini

The 'Gemini' dataset/benchmark is used to evaluate the vulnerability of proprietary Large Language Models (LLMs) to optimization-based prompt injection attacks via their fine-tuning interface.

Papers using Gemini (1)

Fun-tuning: Characterizing The Vulnerability Of Proprietary Llms To Optimization-based Prompt Injection Attacks Via The Fine-tuning Interface2025 · 2 cites