Hi! I am Zhanwen Chen (he/him/his), a Ph.D. student the at the Responsible AI in Varying Environments (RAVE) Lab (PI: Dr. Tom Hartvigsen) at The University of Virginia.

My CV is here (updated February 2, 2025)

Research Interest

I am fascinated by one research question: can computers reason about visual and textual scenes like humans? In other words, how to extract, understand, memorize, and reason with semantic knowledge from both vision and text? I ground this long-term goal to the following research problems:

• How to understand the relationships among objects in scenes (WACV 2023).
• How to answer questions with social/emotional reasoning (ICDL 2020).
• How to enable vision-and-text reasoning in robots.

Research Papers

• Humanizing the Machine: Proxy Attacks to Mislead LLM Detectors

  Tianchun Wang, Yuanzhou Chen, Zichuan Liu, Zhanwen Chen, Haifeng Chen, Xiang Zhang, Wei Cheng

  ICLR 2025

  [Paper]

• Through the Theory of Mind's Eye: Reading Minds with Multimodal Video Large Language Models

  Zhawnen Chen, Tianchun Wang, Yizhou Wang, Michal Kosinski, Xiang Zhang, Yun Fu, Sheng Li

  Under Double-Blind Review

  [Paper]

• More Knowledge, Less Bias: Unbiasing Scene Graph Generation with Explicit Ontological Adjustment

  Zhanwen Chen, Saed Rezayi, Sheng Li

  Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2023, pp. 4023-4032

  [Paper] | [Presentation] | [Poster]

• Characterizing Datasets for Social Visual Question Answering, and the New TinySocial Dataset

  Zhanwen Chen, Shiyao Li, Roxanne Rashedi, Xiaoman Zi, Morgan Elrod-Erickson, Bryan Hollis, Angela Maliakal, Xinyu Shen, Simeng Zhao, Maithilee Kunda

  Joint IEEE International Conference on Development and Learning and on Epigenetic Robotics (ICDL), 2020.

  [Paper] | [Presentation]

• Noise Suppression in Ultrasound Beamforming Using Convolutional Neural Networks

  Zhanwen Chen

  Master's Thesis

  [Paper]

• Compact Convolutional Neural Networks for Ultrasound Beamforming

  Zhanwen Chen, Adam Luchies, Brett Byram

  IEEE International Ultrasonics Symposium (IUS), 2019

  [Paper] | [Code]

Industry Experiences

• Application Developer, Vanderbilt Institute for Clinical and Translational Research
• Full-Stack Web Developer and Data Scientist, Basil Systems Inc.
• Post-Baccalaureate Web Technologist, Five Colleges Inc.
• Software Engineering Intern, Pitney Bowes
• Information Services/Information Technology Intern, Nestle Waters

Awards

• Vanderbilt University Graduate Student Travel Grant to Present Research, 2019
• Inroads Scholarship Award, 2016
• 3rd Place, Area Contest, Toastmasters International Speech Contest

Blog Articles

• Speed Up PyTorch by Building from Source on Ubuntu 18.04
• Install CUDA 10.0 and cuDNN 7.5.0 for PyTorch on Ubuntu 18.04 LTS
• Install OpenCV 4.0.1 from Source on MacOS with Anaconda Python 3.7 to Use SIFT and SURF
• Deploy a Trained RNN/LSTM Model with TensorFlow-Serving and Flask, Part 1: Introduction and Installations

Fun things

I'm a classically trained pianist and a singer (baritone). My favorite composers are J.S. Bach, Sergei Rachmaninoff, and Francis Poulenc. Fun fact: at one point in life I almost became a professional pianist but I decided against the conservatory life. Here's me in concert (I come in at 1:07):

I love hiking and kayaking. I went on my first kayaking trip in Summer 2020 and since then the Vanderbilt ML armada has been terrorizing the rivers of Tennessee. I also enjoy traveling (pre- and hopefully post-COVID) and cooking various cuisines.

My first kayaking trip, 2020

The Vanderbilt ML Armada, 2020

With my girlfriend Angie and our puppy Margeaux in front of the Christmas tree in Franklin, Tennessee, 2020

Roaming the Scottish Lowlands, 2019

Kung Pao Chicken, Lomo Saltado with Aji Verde, Plátanos Maduros, 2020