Scholar
Jiawei Yang
Google Scholar ID: OYrpIa8AAAAJ
University of Southern California
Citations & Impact
Citations
2,105
H-index
15
i10-index
20
Publications
20
Co-authors
8
list available
Publications
Resume (English only)
Academic Achievements
- Publications:
- - Latent Denoising Makes Good Visual Tokenizers (Arxiv 2025)
- - Towards Efficient and Effective Multi-Camera Encoding for End-to-End Driving (Arxiv 2025)
- - STORM: Spatio-temporal reconstruction model for large-scale outdoor scenes (ICLR 2025)
- - Denoising Vision Transformers (ECCV 2024)
- - EmerNeRF: Emergent Spatial-Temporal Scene Decomposition via Self-Supervision (ICLR 2024)
- - FreeNeRF: Improving Few-shot Neural Rendering with Free Frequency Regularization (CVPR 2023)
- - TreeMoCo: Contrastive Neuron Morphology Representation Learning (NeurIPS 2022)
- Awards: Qualcomm Innovation Fellowship (2025); NVIDIA Graduate Fellowship (2025-2026)
Research Experience
- Work Experience: Conducting research on generative models during his PhD at USC; Research Projects: Visual generation and unified multimodal models, representation learning and 3D/4D scene understanding, self-supervised learning in medical and natural images, few-shot NeRF, self-supervised dynamic scene decomposition, etc.
Education
- Degree: CS PhD; School: USC; Advisor: Yue Wang; Year: Third year; Major: Computer Science
Background
- Research Interests: Generative models, visual generation, and unified multimodal models; Professional Field: Computer Science; Summary: Building simple and scalable generative models that span images, video, actions, and unified multimodal systems. The goal is to create a single model that can perceive, understand, and generate the world.
Miscellany
- Personal Interests: Pursuing simplicity and scalability, with the belief that things should be as simple as possible but not simpler.
