generative video · image · safety

I make pixels move.

Amber Yijia Zheng — researcher & builder in generative AI
Ph.D. in Computer Science @ Purdue

scholar github linkedin cv
t = 1.00
noiseclean
/ about

I like fancy visuals — and I build the models that make them.

I recently defended my Ph.D. in Computer Science at Purdue, advised by Dr. Raymond A. Yeh. I received my B.Sc. in Data Science at Shanghai University of Finance and Economics, with Dr. Yixuan Qiu.

✉ let's make something fancy →
/ experience
industry
Research Scientist Intern
Meta Superintelligence Labs
Mar. 2026 - Jun. 2026
Research Scientist Intern
Meta Superintelligence Labs
May 2025 - Jan. 2026
Research Intern
Apple
Apr. 2024 - Sept. 2024
Applied Scientist Intern
Amazon Web Services
Feb. 2022 - Apr. 2024
education
Ph.D. in Computer Science
Purdue University
Aug. 2022 - Aug. 2026
B.sc. in Data Science
Shanghai University of Finance and Economics
Sept. 2018 - Jul. 2022
/ projects
Native Video Generation
MovieGen Team (Core Contributor)
Launched in Meta AI & Instagramshipped
Designing to Forget: Deep Semi-parametric Models for Unlearning
Amber Yijia Zheng*, Yu-Shan Tai*, Raymond A. Yeh
CVPR, 2026Highlight
PaperarXivCode
WebAccessVL: Violation-Aware VLM for Web Accessibility
Amber Yijia Zheng*, Jae Joong Lee*, Bedrich Benes, Raymond A. Yeh
arXiv, 2026
Project PagearXiv
Knowledge Distillation Detection for Open-weights Models
Qin Shi*, Amber Yijia Zheng*, Qifan Song, Raymond A. Yeh
NeurIPS, 2025
PaperarXivCode
DarkDiff: Advancing Low-Light Raw Enhancement by Retasking Diffusion Models for Camera ISP
Amber Yijia Zheng, Yu Zhang, Jun Hu, Raymond A. Yeh, Chen Chen
arXiv, 2025
arXiv
Model Immunization from a Condition Number Perspective
Amber Yijia Zheng*, Site Bai*, Brian Bullins, Raymond A. Yeh
ICML, 2025Oral presentation (Top 1%)
PaperProject PagearXivCode
CFG-Zero*: Improved Classifier-Free Guidance for Flow Matching Models
Weichen Fan, Amber Yijia Zheng, Raymond A. Yeh, Ziwei Liu
arXiv, 2025
Project PagearXivCode
Multi-concept Model Immunization through Differentiable Model Merging
Amber Yijia Zheng, Raymond A. Yeh
AAAI, 2025
PaperProject PagearXivCode
Learning to obstruct few-shot image classification over restricted classes
Amber Yijia Zheng*, Chiao-An Yang*, Raymond A. Yeh
ECCV, 2024
PaperProject PagearXivCode
IMMA: Immunizing text-to-image Models against Malicious Adaptation
Amber Yijia Zheng, Raymond A. Yeh
ECCV, 2024
PaperProject PagearXivCode
Graph Machine Learning through the Lens of Bilevel Optimization
Amber Yijia Zheng, Tong He, Yixuan Qiu, Minjie Wang, David Wipf
AISTATS, 2024
PaperarXivCode
Learning Manifold Dimensions with Conditional Variational Autoencoders
Amber Yijia Zheng, Tong He, Yixuan Qiu, David Wipf
NeurIPS, 2022
PaperarXivCode
/ blogs

No posts yet — first one's in the denoising queue. Coming soon.

/ life

The part of me that isn't a model checkpoint — in rough chronological order. Some of it's exciting, some of it's gloriously boring.

Apr 2026

Defended my Ph.D. 🎓

Defended my Ph.D. 🎓

The committee said Congrats.