L
Lirong Wu
Researcher at Westlake University
Publications - 64
Citations - 385
Lirong Wu is an academic researcher from Westlake University. The author has contributed to research in topics: Computer science & Nonlinear dimensionality reduction. The author has an hindex of 4, co-authored 31 publications receiving 53 citations. Previous affiliations of Lirong Wu include Zhejiang University.
Papers
More filters
Proceedings ArticleDOI
SimGRACE: A Simple Framework for Graph Contrastive Learning without Data Augmentation
TL;DR: A Simple framework for GRAph Contrastive lEarning, SimGRACE, which does not require data augmentations and can yield competitive or better performance compared with state-of-the-art methods in terms of generalizability, transferability and robustness, while enjoying unprecedented degree of flexibility and efficiency.
Proceedings ArticleDOI
Co-learning: Learning from Noisy Labels with Self-supervision
TL;DR: In this paper, the authors proposed a co-learning method for learning with noisy labels, where the intrinsic similarity with the self-supervised module and the structural similarity with noisily supervised module are imposed on a shared common feature encoder to regularize the network to maximize the agreement between the two constraints.
Proceedings ArticleDOI
Co-learning: Learning from Noisy Labels with Self-supervision
TL;DR: In this article, the authors proposed a co-learning method for learning with noisy labels, where the intrinsic similarity with the self-supervised module and the structural similarity with noisily supervised module are imposed on a shared common feature encoder to regularize the network to maximize the agreement between the two constraints.
Proceedings ArticleDOI
A GAN-based Tunable Image Compression System
TL;DR: This paper rethinks content-based compression by using Generative Adversarial Network (GAN) to reconstruct the non-important regions and multiscale pyramid decomposition is applied to both the encoder and the discriminator to achieve global compression of high-resolution images.
Proceedings ArticleDOI
SimVP: Simpler yet Better Video Prediction
TL;DR: This paper proposes SimVp, a simple video prediction model that is completely built upon CNN and trained by MSE loss in an end-to-end fashion that can achieve state-of-the-art performance on five benchmark datasets.