Jintu Zheng 郑晋图

Milestones
I am actively seeking for full-time job opportunities for fall 2025!
I am working as an E2E driving engineer (intern) for ZERON truck (Shanghai, China) now.
I worked as a LLM researcher (intern) for AMD (US) from 2024.02 to 2024.12.
Our work for diffusion speedup was accepted by NeurIPS 2024 Poster.
Our work for 3D structure prediction was accepted by ECCV 2024 Oral.
About me
I am a master student of University of Chinese Academy of Sciences, and major in computer science. I am currently focused on the development of L4 end-to-end systems for autonomous driving and vision language model. I have research experience in driving perception (3D object detection, online hd maps, lane topology) and stand for AMD in the Mapless Driving track at CVPR 2024 AGC, finishing 5th out of 120 teams. I have research experience in large model acceleration, and I am familiar with the deployment and acceleration of large models in practical applications, understand the implementation of custom CUDA and ROCm OPs, and understand basic HPC technologies. This page only shows my public projects (just a small part of all), if you are interest for more, welcome to contact me.
[CV (Chinese)] [Email] [ORCID] [openreview] [dblp]
Resume
Employment
Algorithm Engineer (Inetrn)
2024.12 - Now
Autonmous Driving Department
ZERON Truck, Shanghai China
End-to-end driving, Autolabeling, Data looping, Training speedup
Supervisor: Dr. Panqu Wang
Algorithm Researcher (Inetrn)
2024.02 - 2024.12
AIG-Models
AMD, US
Driving perception, Large language model, Diffusion model
Supervisor: Ji Liu
Education
Master student
2022.09 - Now
Cognitive Interaction Center
Electronic Information, Computer Science
Shenzhen Institute of Advanced Technology
University of Chinese Academy of Sciences, Beijing, China
Supervisor: Prof. Zenan Wang
Undergraduate student
2018.09 - 2022.09
Mathematics and Information
South China Agricultural University, Guangzhou, China
Supervisor: Prof. Yun Liang
Public Projects
My private projects (e.g., autolabeling, LLM/VLM training speedup, better VLM in 3d, etc.), please contact me.

Mapless Driving
Jintu Zheng, Ji Liu, Mingjie Lu; (CVPR Autonmous Grand Challenge 2024) [Rank]

FTP: A Fine-grained Token Pruner for Large Language Models via Token Routing
Jintu Zheng, Zekai Li, Ji Liu, Han Liu, Haowei Zhu, Zeping Li, Fuwei Yang, Haiduo Huang, Jinzhang Peng, Dong Li, Lu Tian, Emad Barsoum;(arXiv, underreview) [Paper]

SparseSSP: 3D Subcellular Structure Prediction from Sparse-View Transmitted Light Images
Jintu Zheng, Yi Ding, Qizhe Liu, Yi Cao, Ying Hu, and Zenan Wang;(ECCV 2024, Oral) [Paper] [Code] [Oral Slide]

Memory Matching is not Enough: Jointly Improving Memory Matching and Decoding for Video Object Segmentation
Jintu Zheng, Yun Liang, Yuqing Zhang, Wanchao Su;(ICPR 2024) [Paper]

SASSM: Semantic Awareness and Self-Support Matching for Semi-Supervised Video Object Segmentation
Yun Liang, Ming Junhui, *Jintu Zheng;(ACM mmasia 2023) [Paper]

A Barcode-Array Biochip Sensor Analysis System Based on Geometry-Guide Learning
Jintu Zheng, Yongting Zhang, Xiaojuan Zhou, Chunhua Wang, Jun Wei, Ying Hu;(IEEE Sensor Journal) [Paper]

HiS4MAE: High-efficiency Segmentation of Subcellular Structure via Self-distillated Masked Autoencoder
Qizhe Liu, Huangxin Xu, Jue HOU, *Jintu Zheng, Zenan Wang;(underreview) [Paper]
Contact
Email:
zhengjintu22@mails.ucas.ac.cn