CAPE is a novel positional encoding that identifies underlying causal structure over non-sequential features to improve the performance of transformer-based models. 
MEATRD is a multimodal anomaly detection method that integrates histology image and Spatial Transcriptomics gene expression data. 
STANDS is a GAN-based multi-task deep learning framework, which can detect and dissect anomalous tissue domains (DDATD) with Spatial Transcriptomics or scRNA-seq. 
ACSleuth is a GAN-based generative model for domain adaptive and fine-grained anomaly detection in the single-cell/tabular data. 