Hua Lu

federated learning (1) representation learning (1) contrastive learning (1) transfer learning (1) knowledge transfer (1) preference alignment (1) dialogue generation (1) reinforcement learning from human feedback (1) hierarchical representation (1) preference modeling (1) human feedback (1) personalized model (1) pre-trained model (1) spatio-temporal datum (1) client drift (1) traffic prediction (1) spatial heterogeneity (1) open-domain dialogue (1) multi-party conversation (1) large language model (1)