Sotabase

Career

· Research Scientist, Meta Superintelligence Labs2025–
· Research Intern, Google DeepMind2024–2025
· Ph.D. in EECS, UC Berkeley2020–

Publications (57)

Long-tailed Recognition by Routing Diverse Distribution-Aware Experts
International Conference on Learning Representations · 2020
462
cited
Cut and Learn for Unsupervised Object Detection and Instance Segmentation
Computer Vision and Pattern Recognition · 2023
241
cited
Debiased Learning from Naturally Imbalanced Pseudo-Labels
Computer Vision and Pattern Recognition · 2022
135
cited
Unsupervised Feature Learning by Cross-Level Instance-Group Discrimination
Computer Vision and Pattern Recognition · 2020
87
cited
A CNN-Based End-to-End Learning Framework Toward Intelligent Communication Systems
IEEE Access · 2019
75
cited
Piezoelectric potential output from ZnO nanowire functionalized with p-type oligomer.
Nano letters (Print) · 2008
75
cited
Clipped Hyperbolic Classifiers Are Super-Hyperbolic Classifiers
Computer Vision and Pattern Recognition · 2021
68
cited
Hierarchical Open-vocabulary Universal Image Segmentation
Neural Information Processing Systems · 2023
63
cited
Unsupervised Hierarchical Semantic Segmentation with Multiview Cosegmentation and Clustering Transformers
Computer Vision and Pattern Recognition · 2022
57
cited
A Novel OFDM Autoencoder Featuring CNN-Based Channel Estimation for Internet of Vessels
IEEE Internet of Things Journal · 2020
53
cited
Unsupervised Visual Attention and Invariance for Reinforcement Learning
Computer Vision and Pattern Recognition · 2021
51
cited
Tied Block Convolution: Leaner and Better CNNs with Shared Thinner Filters
AAAI Conference on Artificial Intelligence · 2020
47
cited
VideoCutLER: Surprisingly Simple Unsupervised Video Instance Segmentation
Computer Vision and Pattern Recognition · 2023
37
cited
Double Insurance: Incentivized Federated Learning with Differential Privacy in Mobile Crowdsensing
IEEE International Symposium on Reliable Distributed Systems · 2020
35
cited
See, Say, and Segment: Teaching LMMs to Overcome False Premises
Computer Vision and Pattern Recognition · 2023
34
cited
Rethinking Patch Dependence for Masked Autoencoders
Trans. Mach. Learn. Res. · 2024
33
cited
The Steady-State Mean-Square Error Analysis for Least Mean p -Order Algorithm
IEEE Signal Processing Letters · 2009
24
cited
Supplementary for Clipped Hyperbolic Classifiers Are Super-Hyperbolic Classifiers
2022
21
cited
Excess MSE analysis of the concurrent constant modulus algorithm and soft decision-directed scheme for blind equalisation
2008
18
cited
Unsupervised Feature Learning by Cross-Level Discrimination between Instances and Groups
arXiv.org · 2020
18
cited
Sotabase