← Search

Zhaolin Zhang

2 accepted papers

2025

DALIP: Distribution Alignment-based Language-Image Pre-Training for Domain-Specific Data

ICCV 2025poster

Recently, Contrastive Language-Image Pre-training (CLIP) has shown promising performance in domain-specific data (e.g., biology), and has attracted increasing research attention. Existing works generally focus on collecting extensive domain-specific data and directly tuning the original CLIP models.…

2022

DropCov: A Simple yet Effective Method for Improving Deep Architectures

NeurIPS 2022accept

Previous works show global covariance pooling (GCP) has great potential to improve deep architectures especially on visual recognition tasks, where post-normalization of GCP plays a very important role in final performance. Although several post-normalization strategies have been studied, these meth…