About me

Hi there! I'm an AI Resident at Meta AI mentored by Satwik Kottur working on multimodal, proactive, goal-oriented, human-augmenting AI. Previously, I completed my undergrad at UMass Amherst, where I was fortunate to have worked with Prof. Mohit Iyyer, Hamed Zamani, and Ina Fiterau on paraphrase generation in video games, multimodal conversational agents, and text-image retrieval respectively. I also previously participated in REU-CAAR at the University of Maryland, where I had the great pleasure of working with Prof. John Dickerson and Tom Goldstein on robustness and bias in facial detection/recognition systems . My research interests lie at the intersection of computer vision, natural language processing, and deep learning (or multimodal machine learning). More specifically in multimodal ML, I am interested in multimodal representation learning, embodied AI, and grounded multimodal generation.


email: gzwei [at] meta [dot] com


  • Sep 2022: One paper got accepted to the NeurIPS Datasets and Benchmarks Track 2022! (link)
  • Jul 2022: I'm joining Meta AI as an AI Resident in Menlo Park, California
  • Jun 2022: I'm serving as a reviewer for the NeurIPS Datasets and Benchmark Track
  • May 2022: Graduated from UMass Amherst
  • Sep 2021: I am participating on UMass Amherst's team for the Alexa Prize Taskbot Challenge
  • Mar 2021: I'm going to be a Student Researcher through REU-CAAR at UMD

© George Zhihong Wei. All Rights Reserved.

Developed by

Last updated: December 2022