Christian Schroeder de Witt

multi-agent system (5) multi-agent reinforcement learning (4) reinforcement learning (3) language model (2) ai safety (2) llm safety (2) decentralized policy (2) large language model (2) adversarial learning (2) steganographic collusion (2) multi-modal learning (1) monotonic optimization (1) zero-shot coordination (1) weather forecasting (1) game theory (1) adversarial defense (1) secret collusion (1) backdoor attack (1) steganographic communication (1) multi-agent deception (1)