Shenglan Li

large language model (2) content generation (2) ai safety (2) copyright infringement (2) self-supervised learning (1) text generation (1) multimodal learning (1) privacy preservation (1) responsible ai (1) neuron model (1) robust learning (1) diffusion model (1) spiking neural network (1) language model (1) gradient inversion attack (1) byte-level tokenization (1) persuasion technique (1) adversarial prompting (1) graph neural network (1) multi-spike learning (1)