Tag
This paper introduces a framework called 'Language Game' that enables dialogue with non-neural biological systems (e.g., gene regulatory networks) by treating communication as a game where the system's internal dynamics are frozen as a reinforcement learning policy, and only linear interfaces are trained. The approach allows diverse systems to communicate fluently without altering their parameters.
This paper introduces a new paradigm for universal Gene Regulatory Network (GRN) inference using single-cell foundation models, proposing Virtual Value Perturbation and Gradient Trajectory methods to distill regulatory knowledge.