Sotabase

Career

· Senior Research Director, Salesforce2022–
· PhD, Computer Science, University at Buffalo2010–2015
· M.Sc., Computer Science, The Hong Kong Polytechnic University2008–2010

Publications (86)

Challenges in representation learning: A report on three machine learning contests
Neural Networks · 2013
1,850
cited
CodeGen: An Open Large Language Model for Code with Multi-Turn Program Synthesis
International Conference on Learning Representations · 2022
1,386
cited
Human Identification Using Finger Images
IEEE Transactions on Image Processing · 2012
606
cited
End-to-End Dense Video Captioning with Masked Transformer
2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2018
581
cited
Learn to Grow: A Continual Structure Learning Framework for Overcoming Catastrophic Forgetting
International Conference on Machine Learning · 2019
498
cited
Human Identification Using Palm-Vein Images
IEEE Transactions on Information Forensics and Security · 2011
298
cited
CodeGen2: Lessons for Training LLMs on Programming and Natural Languages
arXiv.org · 2023
223
cited
UniControl: A Unified Diffusion Model for Controllable Visual Generation In the Wild
Neural Information Processing Systems · 2023
201
cited
A Conversational Paradigm for Program Synthesis
arXiv.org · 2022
189
cited
LLMs Get Lost In Multi-Turn Conversation
arXiv.org · 2025
134
cited
Normalization Propagation: A Parametric Technique for Removing Internal Covariate Shift in Deep Networks
International Conference on Machine Learning · 2016
131
cited
Human identification using KnuckleCodes
2009 IEEE 3rd International Conference on Biometrics: Theory, Applications, and Systems · 2009
103
cited
Personal identification using finger knuckle orientation features
Electronics Letters · 2009
103
cited
Why Regularized Auto-Encoders learn Sparse Representation?
International Conference on Machine Learning · 2015
69
cited
Long Document Summarization with Top-down and Bottom-up Inference
Findings · 2022
67
cited
CoCo: Controllable Counterfactuals for Evaluating Dialogue State Trackers
International Conference on Learning Representations · 2020
65
cited
Parallel Feature Selection Inspired by Group Testing
Neural Information Processing Systems · 2014
61
cited
Contactless palm vein identification using multiple representations
International Conference on Biometrics: Theory, Applications and Systems · 2010
60
cited
Contactless fingerprint identification using level zero features
CVPR 2011 WORKSHOPS · 2011
59
cited
A Multi-Discriminator CycleGAN for Unsupervised Non-Parallel Speech Domain Adaptation
Interspeech · 2018
56
cited
Sotabase