Xinyan Guan

Reinforcement Learning (1)

large language model (5) hallucination mitigation (2) instruction tuning (2) preference learning (1) prompt engineering (1) document summarization (1) fact verification (1) text representation (1) synthetic data generation (1) latent space (1) reinforcement learning from human feedback (1) knowledge graph (1) language model (1) reward model (1) parameter-efficient fine-tuning (1) dialogue generation (1) self-supervised learning (1) on-policy learning (1) sparse autoencoder (1) continual learning (1)