About Me

Hi! I am Shijia Yang, a second-year Master's student in Computer Science at Stanford University. I was advised by Professor Kurt Keutzer at BAIR Lab and Professor Masayoshi Tomizuka at MSC Lab.

My goal is to extend human intelligence and physical power beyond the limit. Therefore, I hope to build a robust and intelligent robotic perception system. My specific interest is to explore methods that properly and efficiently utilizing multimodal data, including image, language, point-cloud, and high dimension variant of them, to create realiable representations for machine perception. I also love to explore other interesting directions.

News: I am currently researching on vision-language alignment. Actively looking for communication and collaboration!

Publications

Law of Vision Representation in MLLMs

Shijia Yang, Bohan Zhai, Quanzeng You, Jianbo Yuan, Hongxia Yang, Chenfeng Xu.

• Paper
• Code

HallE-Control: Controlling Object Hallucination in Large Vision Language Models

Shijia Yang*, Bohan Zhai*, Chenfeng Xu, Sheng Shen, Kurt Keutzer, Manling Li.

• Paper
• Code

Multitask Vision-Language Prompt Tuning

Shijia Yang*, Sheng Shen*, Tianjun Zhang*, Bohan Zhai, Joseph E. Gonzalez, Kurt Keutzer, Trevor Darrell.

WACV 2024.

• Paper
• Code

Time Will Tell: New Outlooks and A Baseline for Temporal Multi-view 3D Object Detection

Jinhyung Park*, Chenfeng Xu*, Shijia Yang, Kurt Keutzer, Kris Kitani, Masayoshi Tomizuka, Wei Zhan.

ICLR 2023 Oral.

• Paper
• Code

Image2Point: 3D Point-Cloud Understanding with 2D Image Pretrained Models

Shijia Yang*, Chenfeng Xu*, Tomer Galanti, Bichen Wu, Xiangyu Yue, Bohan Zhai, Wei Zhan, Peter Vajda, Kurt Keutzer, Masayoshi Tomizuka.

ECCV 2022.

• Paper
• Code

Curriculum Vitae