Mingyang Wan

large language model (2) transfer learning (2) curriculum learning (1) knowledge distillation (1) multimodal learning (1) instruction tuning (1) reinforcement learning from human feedback (1) image super-resolution (1) monte carlo tree search (1) low-resource language (1) language model (1) instruction fine-tuning (1) mechanistic interpretability (1) sparse autoencoder (1) weak-to-strong generalization (1) inference-time alignment (1) pretrained model (1) information injection (1) dialogue system (1) hyperspectral image (1)