Xiangzhi Tong

I am currently an M.Eng. student @ Cornell University in Ithaca, NY.

I used to be a fulltime Software Engineer @ Zoom.

Email  /  Google Scholar  /  LinkedIn  /  Github

profile photo

🎓 Education

[Cornell University], M.Eng in Electrical and Computer Engineering, 2023-2024

[Xi'an Jiaotong-Liverpool University], B.Eng in Computer Science, 2018-2022

[University of Liverpool], B.Eng in Computer Science, 2018-2022

🗞️ News

[2026.5] ViViD-5K, our vineyard vision dataset, and the GrapeSAM pipeline are out — project page / arXiv.

🗿 Research

I'm passionate about applying vision-language models in robotics. Representative papers are highlighted.

GrapeSAM berry segmentation example
ViViD-5K: Vineyard vision dataset for field-based berry detection and segmentation and grape cluster closure estimation
Xiangzhi Tong, Chengrui Zhang, Mac Flaherty, Andre Matteo Garcia, Dominic Gorman, Jonathan Jaramillo, Justine E. Vanden Heuvel, Yu Jiang
arXiv preprint, 2026
project page / arXiv / code / dataset

GrapeSAM: a unified in-field pipeline for grape cluster segmentation, berry detection and counting, and visual cluster closure estimation.

SHARP overview: software hints select phase-specific memory access prediction models
SHARP: Software Hint-Assisted Memory Access Prediction for Graph Analytics
Pengmiao Zhang, Rajgopal Kannan, Xiangzhi Tong, Anant V. Nori, Viktor K. Prasanna
HPEC, 2022
paper

Programmer-inserted software hints expose the Scatter/Gather phase of graph workloads to phase-specific attention models, improving memory access prediction F1 by 9%–22% over Delta-LSTM and 3%–8% over TransFetch.

Predicted urban crime occurrence map
Urban crime records map
Urban Crime Trends Analysis and Occurrence Possibility Prediction based on Light Gradient Boosting Machine
Xiangzhi Tong, Pin Ni, Qingge Li, Qiao Yuan, Junru Liu, Hanzhe Lu, Gangmin Li
BDAI, 2021
paper

LightGBM trained on real urban crime records predicts the possibility of crime occurrence, with predictions closely tracking actual crime variation — a step toward data-driven crime prevention and policing.

Miscellanea
Nothing......Yet!
Liv_LOGO 2020-2021 University Academic Achievement Award (Top 6%)
2018-2019 University Academic Achievement Award (Top 10%)

Blog Posts

 Nothing.....Yet

Feel free to steal this website's source code.