Zhemin Zhang

2 accepted papers

2025

Efficient Visual Representation Learning with Heat Conduction Equation

IJCAI 2025

Foundation models, such as CNNs and ViTs, have powered the development of image representation learning. However, general guidance to model architecture design is still missing. Inspired by the connection between image representation learning and heat conduction, we model images by the heat conducti

Cited by 0 Source ↗PDF ↗code ↗Scholar ↗

2023

Positional Label for Self-Supervised Vision Transformer

Zhemin Zhang, Xun Gong

AAAI 2023technical

Positional encoding is important for vision transformer (ViT) to capture the spatial structure of the input image. General effectiveness has been proven in ViT. In our work we propose to train ViT to recognize the positional label of patches of the input image, this apparently simple task actually y…

Cited by 8 Source ↗PDF ↗code ↗Scholar ↗