Defending Jailbreak Prompts via In-Context Adversarial Game (ICAG)
University of Notre Dame, INRIA, King Abdullah University of Science and Technology (2024)
https://aclanthology.org/2024.emnlp-main.1121.pdf
Метод поэтапной генерации системного промпта как игры между двумя LLM (attack agent и defense agent)
Post #1572
240


