Tag
This paper presents CARL, an autotelic reinforcement learning framework that autonomously discovers and controls self-organizing phenomena in complex systems like Lenia cellular automata, demonstrating capabilities in steering solitons and zero-shot generalization.
This paper proposes Semantic Lenia, a framework that transforms LLM inference into a continuous dynamical system in logit space, demonstrating the emergence of autonomous semantic solitons and homeostatic limit cycles through nonlinear feedback.