sparse activation

mixture of expert (30) model compression (12) parameter efficiency (9) large language model (7) neural network (6) language model (5) feed-forward network (4) expert routing (4) load balancing (4) continual learning (4)