Yuto Harada

representation learning (1) information bottleneck (1) preference learning (1) multimodal learning (1) language model alignment (1) instruction tuning (1) model alignment (1) mutual information (1) language acquisition (1) layer-wise analysis (1) human feedback (1) recurrent neural network (1) model training (1) supervised fine-tuning (1) morphological inflection (1) reading time (1) llm alignment (1) buzzer quiz (1) bidirectional learning (1) large language model (1)