Ke Zeng

large language model (8) reinforcement learning (6) policy optimization (3) catastrophic forgetting (1) anomaly detection (1) curriculum learning (1) knowledge transfer (1) preference learning (1) information gain (1) knowledge distillation (1) instruction following (1) paraphrase generation (1) computational efficiency (1) instruction tuning (1) importance sampling (1) reinforcement learning from human feedback (1) event detection (1) adaptive computation (1) mathematical reasoning (1) multi-task learning (1)