Sotabase

Career

· Assistant Professor, Rutgers University2025–
· Research Scientist, Google2024–2025
· Assistant Professor, McGill University2023–2024
· Core Faculty Member, MILA - Quebec Artificial Intelligence Institute2023–2024
· PhD Student, Columbia University2018–2023
· Bachelor's degree, Tsinghua University2013–2018
· Visiting Student, MIT

Publications (43)

Metric Learning for Adversarial Robustness
Neural Information Processing Systems · 2019
202
cited
Understanding Zero-Shot Adversarial Robustness for Large-Scale Models
International Conference on Learning Representations · 2022
113
cited
Generative Interventions for Causal Learning
Computer Vision and Pattern Recognition · 2020
70
cited
Adversarial Attacks are Reversible with Natural Supervision
IEEE International Conference on Computer Vision · 2021
65
cited
Raidar: geneRative AI Detection viA Rewriting
International Conference on Learning Representations · 2024
56
cited
SelfIE: Self-Interpretation of Large Language Model Embeddings
International Conference on Machine Learning · 2024
48
cited
Discrete Representations Strengthen Vision Transformer Robustness
International Conference on Learning Representations · 2021
46
cited
Causal Transportability for Visual Recognition
Computer Vision and Pattern Recognition · 2022
41
cited
Towards Causal Deep Learning for Vulnerability Detection
International Conference on Software Engineering · 2023
37
cited
Doubly Right Object Recognition: A Why Prompt for Visual Rationales
Computer Vision and Pattern Recognition · 2022
35
cited
A Probabilistic Learning Approach to UWB Ranging Error Mitigation
Global Communications Conference · 2018
32
cited
Live Trojan Attacks on Deep Neural Networks
2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · 2020
25
cited
Bidirectional Inference Networks: A Class of Deep Bayesian Networks for Health Profiling
AAAI Conference on Artificial Intelligence · 2019
24
cited
Convolutional Visual Prompt for Robust Visual Perception
Neural Information Processing Systems · 2023
20
cited
AdvSPADE: Realistic Unrestricted Attacks for Semantic Segmentation
2019
14
cited
I Can Hear You: Selective Robust Training for Deepfake Audio Detection
International Conference on Learning Representations · 2024
10
cited
Robust Perception through Equivariance
International Conference on Machine Learning · 2022
10
cited
Shadows Shed Light on 3D Objects
arXiv.org · 2022
10
cited
RAFT: Realistic Attacks to Fool Text Detectors
Conference on Empirical Methods in Natural Language Processing · 2024
9
cited
Turns Out I'm Not Real: Towards Robust Detection of AI-Generated Videos
arXiv.org · 2024
9
cited
Sotabase
Chengzhi Mao | Researcher Profile | Sotabase | Sotabase