Cyril Zhang

Microsoft (United States) — United States

Computer Science · Physical Sciences

12h-index406citations47works18.72yr avg
OpenAlex↗ORCID↗Google Scholar↗Lab page↗

Accepting Students?

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

Research Topics

Machine Learning and Algorithms(10), Advanced Bandit Algorithms Research(8), Topic Modeling(6), Model Reduction and Neural Networks(6), Natural Language Processing Techniques(5)

Publications47 total

EMA Without the Lag: Bias-Corrected Iterate Averaging Schemes
arXiv (Cornell University)·2025Open Access
EyeO: Autocalibrating Gaze Output with Gaze Input for Gaze Typing
2025
Phi-4 Technical Report
arXiv (Cornell University)·2024· 19 citedOpen Access
Self-Improvement in Language Models: The Sharpening Mechanism
arXiv (Cornell University)·2024· 1 citedOpen Access
ASL STEM Wiki: Dataset and Benchmark for Interpreting STEM Articles
arXiv (Cornell University)·2024Open Access
Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone
arXiv (Cornell University)·2024· 143 citedOpen Access
Can large language models explore in-context?
arXiv (Cornell University)·2024· 2 citedOpen Access
ASL STEM Wiki: Dataset and Benchmark for Interpreting STEM Articles
2024· 3 citedOpen Access
Can large language models explore in-context?
2024
Butterfly Effects of SGD Noise: Error Amplification in Behavior Cloning and Autoregression
arXiv (Cornell University)·2023Open Access
Understanding and Improving Neural Active Learning on Heteroskedastic Distributions
Frontiers in artificial intelligence and applications·2023Open Access
Pareto Frontiers in Neural Feature Learning: Data, Compute, Width, and Luck
arXiv (Cornell University)·2023Open Access
EyeO: Autocalibrating Gaze Output with Gaze Input
arXiv (Cornell University)·2023Open Access
Exposing Attention Glitches with Flip-Flop Language Modeling
arXiv (Cornell University)·2023· 8 citedOpen Access
Learning Hidden Markov Models Using Conditional Samples
arXiv (Cornell University)·2023Open Access
Neural Active Learning on Heteroskedastic Distributions
arXiv (Cornell University)·2022Open Access
Transformers Learn Shortcuts to Automata
arXiv (Cornell University)·2022· 11 citedOpen Access
Recurrent Convolutional Neural Networks Learn Succinct Learning Algorithms
arXiv (Cornell University)·2022Open Access
Hidden Progress in Deep Learning: SGD Learns Parities Near the Computational Limit
arXiv (Cornell University)·2022· 12 citedOpen Access
Understanding Contrastive Learning Requires Incorporating Inductive Biases
arXiv (Cornell University)·2022· 15 citedOpen Access
page 1 of 3Next →

Frequent Co-authors

Elad Hazan(14), Naman Agarwal(11), Sham M. Kakade(10), Akshay Krishnamurthy(10), Surbhi Goel(9), Jordan T. Ash(7), Holden Lee(6), Karan Singh(6), Udaya Ghai(6), Karan Singh(5), Daniel Suo(4), Paula Gradu(4), Xinyi Chen(4), Danielle Bragg(4), Dylan J. Foster(4), Yi Zhang(3), Benjamin Edelman(3), Brian Bullins(3), Edgar Minasyan(3), Julienne LaChance(3)