Simple and Effective Masked Langauge Diffusion Models
While diffusion models excel at generating high-quality images, they have traditionally lagged behind autoregressive (AR) methods in language modeling. We ap...
Projects
Joint Analysis of Single-Cell Data across Cohorts with Missing Modalities
Joint analysis of multi-omic single-cell data across cohorts has significantly enhanced the comprehensive analysis of cellular processes. However, most of th...
Hybrid Geometric Primitive Representations for Large Scale Point Clouds
Point clouds, collected by LiDAR scanners and other 3D sensing devices, are commonly used by deep learning models to perceive the physical world. However, th...
Disentangling Multi-Scale Anomaly Signals in Networks via Scale-Aware Contrastive Learning
Detecting suspicious or unexpected patterns in networks without supervision is challenging because it requires heuristic knowledge about anomaly characterist...
Identifying Optimal Proteins by Their Structure Using Graph Neural Networks
While standard protein representations depend on sequence, we hypothesize that protein structure provides richer information for the model to learn these pro...