@online{huangndmodel,
  author       = {Huang, Xuanqiang Angelo},
  title        = {Model Distillation},
  organization = {Xuanqiang Angelo Huang's Blog},
  url          = {https://flecart.github.io/notes/model-distillation/},
  langid       = {english},
  abstract     = {Model Distillation \# Distillation trains a small student q θ ​ to reproduce the behaviour of a large teacher p . Here we revise it for language models: what the loss is, which divergence to use, where the training text should come from, what you can do with only API access, and w}
}