Haoling Li

large language model (2) model compression (1) neural network pruning (1) continual learning (1) catastrophic forgetting (1) image generation (1) domain adaptation (1) gradient-based optimization (1) preference learning (1) direct preference optimization (1) code generation (1) flow matching (1) parameter selection (1) neural network optimization (1) instruction tuning (1) diffusion model (1) language model (1) scaling law (1) parameter-efficient tuning (1) sparse training (1)