Colin Wei

Computer Science · Physical Sciences

17h-index1.0kcitations33works30.02yr avg
OpenAlex↗Google Scholar↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Domain Adaptation and Few-Shot Learning(10), Advanced Neural Network Applications(8), Stochastic Gradient Optimization Techniques(8), Machine Learning and Algorithms(8), Machine Learning and ELM(8)

Publications33 total

OpenAI GPT-5 System Card
arXiv (Cornell University)·2025· 2 citedOpen Access
OpenAI GPT-5 System Card
ArXiv.org·2025Open Access
GPT-4o System Card
arXiv (Cornell University)·2024· 88 citedOpen Access
Max-Margin Works while Large Margin Fails: Generalization without Uniform Convergence
arXiv (Cornell University)·2022Open Access
Beyond Separability: Analyzing the Linear Transferability of Contrastive Representations to Related Subpopulations
arXiv (Cornell University)·2022· 2 citedOpen Access
Provable Guarantees for Self-Supervised Deep Learning with Spectral Contrastive Loss
arXiv (Cornell University)·2021· 6 citedOpen Access
Why Do Pretrained Language Models Help in Downstream Tasks? An Analysis of Head and Prompt Tuning
arXiv (Cornell University)·2021Open Access
Statistically Meaningful Approximation: a Case Study on Approximating Turing Machines with Transformers
arXiv (Cornell University)·2021· 6 citedOpen Access
Why Do Pretrained Language Models Help in Downstream Tasks? An Analysis of Head and Prompt Tuning
arXiv (Cornell University)·2021· 33 citedOpen Access
Provable Guarantees for Self-Supervised Deep Learning with Spectral Contrastive Loss
arXiv (Cornell University)·2021· 36 citedOpen Access
Meta-learning Transferable Representations with a Single Target Domain
arXiv (Cornell University)·2020· 1 citedOpen Access
Theoretical Analysis of Self-Training with Deep Networks on Unlabeled Data
arXiv (Cornell University)·2020· 74 citedOpen Access
Theoretical Analysis of Self-Training with Deep Networks on Unlabeled Data
arXiv (Cornell University)·2020· 20 citedOpen Access
Self-training Avoids Using Spurious Features Under Domain Shift
arXiv (Cornell University)·2020· 5 citedOpen Access
Shape Matters: Understanding the Implicit Bias of the Noise Covariance
arXiv (Cornell University)·2020· 2 citedOpen Access
Shape Matters: Understanding the Implicit Bias of the Noise Covariance
arXiv (Cornell University)·2020· 19 citedOpen Access
Improved Sample Complexities for Deep Neural Networks and Robust Classification via an All-Layer Margin
arXiv (Cornell University)·2020· 11 citedOpen Access
The Implicit and Explicit Regularization Effects of Dropout
arXiv (Cornell University)·2020· 27 citedOpen Access
Improved Sample Complexities for Deep Networks and Robust Classification via an All-Layer Margin
arXiv (Cornell University)·2019· 28 citedOpen Access
Regularization Matters: Generalization and Optimization of Neural Nets v.s. their Induced Kernel
neural information processing systems·2019· 36 cited
page 1 of 2Next →

Frequent Co-authors

Tengyu Ma(21), Jeff Z. HaoChen(6), Jason D. Lee(5), Yining Chen(4), Adrien Gaidon(3), AJ Ostrow(3), Alex Baker-Whitcomb(3), Alexi Christakis(3), Ananya Kumar(3), Brad Lightcap(3), Brian Hsu(3), Camillo Lugaresi(3), Chelsea Voss(3), Chris Hallacy(3), Coley Czarnecki(3), Tengyu Ma(3), Yuanzhi Li(2), Kendrick Shen(2), Sang Michael Xie(2), Iain Murray(2)