Jiawei Yang

CS PhD at USC


I’m currently a first-year PhD student at USC, supervised by Prof. Yue Wang (He is super nice!)

My research interests lie in the intersection of computer vision, computer graphics, and deep learning. Now, I’m also exploring the possibility of language models. In particular, my interest centers around representation learning, spanning from perception to reasoning, reconstruction, and generation. I’d like to bridge the gap between human cognition and machine intelligence.

“Make things as simple as possible, but not simpler.”

At the heart of my research philosophy is the power of simplicity and scalability. I am drawn to ideas that are elegantly simple yet have the potential to scale effectively and make a significant impact.

News

May 03, 2024 One Parallelized Spatiotemporal Binding was accepted to ICML 2024! We developed a parallelized architecture for learning slot representations from sequential inputs.
Jan 16, 2024 Our EmerNeRF is accepted to ICLR2024! It’s a fully self-supervised neural scene representation designed for challenging real-world dynamic driving scenes!
Jan 08, 2024 Super excited to join Google Research as a student researcher, supervised by Yonglong Tian! 🥳
Jan 06, 2024 Check out our Denoising Vision Transformer, which identifies and addresses the noise artifacts inherent in Vision Transformers!
Aug 18, 2023 I formally start my PhD at USC.
Jul 10, 2023 Join Autonomous Vehicle Group at Nvidia Research as a research intern.
Feb 27, 2023 Our FreeNeRF was accepted to CVPR 2023. We can address few-shot neural rendering issues just by a few lines of code.
Sep 14, 2022 One paper on Neuron Morphology Representation Learning was accepted to NeurIPS 2022.
Jul 08, 2022 One paper on self-supervised pre-training for dense prediction was accepted to ECCV 2022.
Jun 02, 2022 Two papers were accepted to MICCAI 2022.

Selected Publications

  1. arxiv
    Denoising Vision Transformers
    Jiawei Yang*, Katie Z Luo*, Jiefeng Li, Kilian Q Weinberger, Yonglong Tian, and Yue Wang
    arXiv preprint arXiv:2401.02957 (arxiv), 2024
  2. ICLR
    EmerNeRF: Emergent Spatial-Temporal Scene Decomposition via Self-Supervision
    Jiawei Yang, Boris Ivanovic, Or Litany, Xinshuo Weng, Seung Wook Kim, Boyi Li, Tong Che, Danfei Xu, Sanja Fidler, Marco Pavone, and Yue Wang
    International Conference on Learning Representations (ICLR), 2024
  3. CVPR
    FreeNeRF: Improving Few-shot Neural Rendering with Free Frequency Regularization
    Jiawei Yang, Marco Pavone, and Yue Wang
    Proceedings of the IEEE/CVF International Conference on Computer Vision (CVPR), 2023
  4. NeurIPS
    TreeMoCo: Contrastive Neuron Morphology Representation Learning
    Hanbo Chen*, Jiawei Yang*, Daniel Iascone, Lijuan Liu, Lei He, Hanchuan Peng, and Jianhua Yao
    Advances in Neural Information Processing Systems (NeurIPS), 2022
  5. ECCV
    ConCL: Concept Contrastive Learning for Dense Prediction Pre-training in Pathology Images
    Jiawei Yang, Hanbo Chen, Yuan Liang, Junzhou Huang, Lei He, and Jianhua Yao
    European Conference on Computer Vision (ECCV), 2022
  6. ICLR
    Towards Better Understanding and Better Generalization of Low-shot Classification in Histology Images with Contrastive Learning
    Jiawei Yang, Hanbo Chen, Jiangpeng Yan, Xiaoyu Chen, and Jianhua Yao
    International Conference on Learning Representations (ICLR), 2022