From Word Models to World Models: Training AI for Adversarial Robustness
238
Shift LLMs from next-token to next-state prediction by training in multi-agent, hidden-state environments so their outputs survive adversarial adaptation.
Game Theory
Strategic decision-making frameworks including Nash equilibria, imperfect information games, multi-agent reasoning, and theory of mind as applied to AI systems and real-world adversarial environments.
Reading List