GAO, Xinyue. Safety Alignment of Large Language Models: The Offensive-Defensive Game Mechanism under Multi-round Adversarial Prompting. CPS Digital Library - Series of Conferences, [S. l.], v. 1, p. 33–37, 2026. Disponível em: https://seriesofconference.com/index.php/SCJ/article/view/266. Acesso em: 31 aug. 2026.