Input Perturbation Reduces Exposure Bias in Diffusion Models

方法与贡献

研究扩散模型采样过程中累积的错误，并在训练中扰动输入，以模拟推理时的预测误差。

通过合成输入扰动处理同一种训练与采样不匹配问题；Scheduled Sampling 则在模型生成的状态上训练。

结合主题阅读

学习渲染 LaTeX、SMILES、HTML 与乐谱 →

引用

Mang Ning, Enver Sangineto, Angelo Porrello, Simone Calderara, Rita Cucchiara. Input Perturbation Reduces Exposure Bias in Diffusion Models. ICML 2023.

@inproceedings{pmlr-v202-ning23a,
  title = {{Input Perturbation Reduces Exposure Bias in Diffusion Models}},
  author = {Ning, Mang and Sangineto, Enver and Porrello, Angelo and Calderara, Simone and Cucchiara, Rita},
  year = {2023},
  booktitle = {ICML 2023},
  url = {https://proceedings.mlr.press/v202/ning23a.html}
}