搜张老师的Google Scholar，筛掉了一些不感兴趣的文章，找到了这一篇蛮有意思的文章，关于人脸编辑的文章，通过语义解构解决了人脸上属性之间的纠缠的挑战，这里挂一个arxiv的链接
2309.05314 Semantic Latent Decomposition with Normalizing Flows for Face Editing (arxiv.org)
However, the resulting methods usually encounter challenges in complicated navigation due to the entanglement among different attributes in the latent space.
在潜在空间中不同属性之间的纠缠
读abstract的时候积累到一个好词组qualitatively and quantitatively以后在写论文的时候，吹自己论文怎么怎么牛逼的时候就这么写’our model outperforms the state of the art model qualitatively and quantitatively’。
回归正题，为了解决上述问题，提出了一个全新的模型
a novel framework, termed SDFlow, with a semantic decomposition in original latent space using continuous conditional normalizing flows
怎么个解决法？
有两个子模块
