Liangyu Huo

reinforcement learning (2) reward modeling (1) few-shot learning (1) imitation learning (1) transfer learning (1) curriculum learning (1) preference learning (1) model architecture (1) dynamic routing (1) adaptive sampling (1) reinforcement learning from human feedback (1) reward function (1) preference modeling (1) adversarial defense (1) bias mitigation (1) mixture of expert (1) multi-objective reinforcement learning (1) feed-forward network (1) pre-trained language model (1) feedforward network (1)