WenXin Dong

Hi! Welcome to this corner of the internet :) My name is WenXin Dong. I joined Google as a software engineer after graduating from Stanford University in 2023 with a BS and a MS degree in CS. During college, I conducted AI research in the areas of reinforcement learning, NLP, computer vision, and graph learning. I worked as research assistant at Stanford Center for Ocean Solutions, and was an undergraduate researcher at Professor Noah Goodman and Professor Leonidas Guibas lab. In the past, I have published a Python book and co-founded a start up project named Club Cardinal. In my free time, I like to draw and paint :) Check out my research and side projects!

Last updated: Feb 13 2024

Frontiers 2023

Comparing spatial patterns of marine vessels between vessel-tracking data and satellite imagery
Shinnosuke Nakayama, WenXin Dong, Richard G. Correro, Elizabeth R. Selig, Colette C.C. Wabnitz, Trevor J. Hastie, Jim Leape, Serena Yeung and Fiorenza Micheli


NeurIPS 2021

Contrastive Reinforcement Learning of Symbolic Reasoning Domains
Gabriel Poesia, WenXin Dong, Noah Goodman


Electronic Industry Press

The Python Algorithm Book that Even You Can Understand
Shuo Wang, WenXin Dong, ShuHang Zhang, Jie Zhang

*Sold over 30,000 copies since 2019

[ISBN-13: 9787121352553]


Using Temporal Contrast to Detect Small Vessels in Low Resolution Optical Satellite Imagery

Detect small vessels in low resolution satellite imagery using temporal fusion in Faster R-CNN



Making Deep Knowledge Tracing Interpretable Using Distillation

Distill transformer-based deep knowledge tracing models into a similarly accurate soft decision tree with interpretable features



Solving the Rubiks’ Cube with Deep Reinforcement Learning using Multi-Headed Models

Using reinforcement learning and multi-task learning to solve the Rubik's cube layer by layer.



Keywords to Story: Controllable Text Generation

Generating 5-sentence English stories which contains specified input keywords.



Detecting the Emotion Dynamics of Profanity Language In Social Media Through Time

- NLP emotional analysis on tweets containing curse words to identify changes in emotional use of curse words from 2011 to 2019.


During the COVID-19 pandemic, 4 Stanford students and I built a website called Club Cardinal, a virtual Stanford campus. After seeing its success, we wanted to build a virtual world available for everyone to use: Virtuali.

More coming soon! :)