Fred Zhang

z0@berkeley.edu

I am an AI research scientist at Meta. Previously, I was at Google DeepMind, working on post-training & RL, where I contributed to Gemini 2.5, Gemini 3.0, Gemini 3.1, IMO '25🥇, ICPC '25🥇 and AlphaProof.

I completed my Ph.D. from the Theory Group at Berkeley EECS in 2024, advised by Jelani Nelson. Prior to that, I received B.S. in Computer Science and B.S. in Mathematics from Duke University.

Links: Google Scholar / Twitter / LinkedIn


Publications show recent / by date / by topic

Topics: Language model / Algorithmic statistics / Sublinear algorithms / Learning-based algorithms / Others

Formal Conjectures: An Open and Evolving Benchmark for Verified Discovery in Mathematics
Moritz Firsching*, Paul Lezeau*, Salvatore Mercuri*, Miklós Z. Horváth*, Yaël Dillies, Calle Sönne, Eric Wieser, Fred Zhang, Thomas Hubert, Blaise Agüera y Arcas, Pushmeet Kohli.

arXiv, 2026

Gemini achieves gold-medal level at the International Collegiate Programming Contest World Finals
Google DeepMind, ICPC 2025 Team (incl. Fred Zhang).

Blogpost, 2025

Advanced version of Gemini with Deep Think officially achieves gold-medal standard at the International Mathematical Olympiad
Google DeepMind, IMO 2025 Team (incl. Fred Zhang as key contributor).

Blogpost, 2025 (media coverage: NYT, WSJ, Ars Technica, Reuters)

Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
Gemini Team (incl. Fred Zhang).

Technical report, 2025 (arXiv)

ForecastBench: A Dynamic Benchmark of AI Forecasting Capabilities
Ezra Karger*, Houtan Bastani*, Chen Yueh-Han*, Zachary Jacobs, Danny Halawi, Fred Zhang, Philip E. Tetlock (* Equal contribution).

ICLR 2025 (arXiv, code)

Approaching Human-Level Forecasting with Language Models
Danny Halawi*, Fred Zhang*, Chen Yueh-Han*, Jacob Steinhardt (* Equal contribution).

NeurIPS 2024 (arXiv, Twitter thread, blogpost, code)

Towards Best Practices of Activation Patching in Language Models: Metrics and Methods
Fred Zhang, Neel Nanda (contribution order).

ICLR 2024 (arXiv, Twitter thread)

Adaptive Regret for Bandits Made Possible: Two Queries Suffice
Zhou Lu, Qiuyi Zhang, Xinyi Chen, Fred Zhang, David P. Woodruff, Elad Hazan (contribution order)

ICLR 2024 (arXiv)

Constant Approximation for Individual Preference Stable Clustering
Anders Aamand, Justin Y. Chen, Allen Liu, Sandeep Silwal, Pattara Sukprasert, Ali Vakilian, Fred Zhang (alphabetical order)

NeurIPS 2023  Spotlight

On Robust Streaming for Learning with Experts: Algorithms and Lower Bounds
David P. Woodruff, Samson Zhou, Fred Zhang (alphabetical order).

NeurIPS 2023 (arXiv)

Privately Estimating a Gaussian: Efficient, Robust and Optimal
Daniel Alabi, Pravesh K. Kothari, Pranay Tankala, Prayaag Venkat, Fred Zhang (alphabetical order).

STOC 2023 (arXiv)

Bayesian Strategy-Proof Facility Location via Robust Estimation
Manolis Zampetakis, Fred Zhang (alphabetical order).

AISTATS 2023 (Paper)

Robust Algorithms on Adaptive Inputs from Bounded Adversaries
Yeshwanth Cherapanamjeri, Sandeep Silwal, David P. Woodruff, Qiuyi Zhang, Fred Zhang, Samson Zhou (alphabetical order).

ICLR 2023 (arXiv)

Online Prediction in Sub-linear Space
Binghui Peng, Fred Zhang (alphabetical order).

SODA 2023  Best Student Paper  (arXiv, talk at Google, talk at TTIC)

Optimal Query Complexities for Dynamic Trace Estimation
David P. Woodruff, Qiuyi Zhang, Fred Zhang (alphabetical order)

NeurIPS 2022 (arXiv)

Faster Fundamental Graph Algorithms via Learned Predictions
Justin Y. Chen, Sandeep Silwal, Ali Vakilian, Fred Zhang (alphabetical order).

ICML 2022 (arXiv)

Robust and Heavy-Tailed Mean Estimation Made Simple, via Regret Minimization
Samuel B. Hopkins, Jerry Li, Fred Zhang (alphabetical order).

NeurIPS 2020 (arXiv)

Optimal Robustness-Consistency Trade-offs for Learning-Augmented Online Algorithms
Alexander Wei, Fred Zhang (alphabetical order).

NeurIPS 2020 (arXiv)

A Fast Spectral Algorithm for Mean Estimation with Sub-Gaussian Rates
Zhixian Lei, Kyle Luh, Prayaag Venkat, Fred Zhang (alphabetical order).

COLT 2020 (arXiv)

SGD on Neural Networks Learns Functions of Increasing Complexity
Preetum Nakkiran, Gal Kaplun, Dimitris Kalimeris, Tristan Yang, Benjamin L. Edelman, Fred Zhang, Boaz Barak (contribution order).

NeurIPS 2019  Spotlight   ( arXiv)

Minimum Cut and Minimum k-Cut in Hypergraphs via Branching Contractions
Kyle Fox, Debmalya Panigrahi, Fred Zhang (alphabetical order).

SODA 2019 & ACM Transactions on Algorithms (TALG) (journal version)


Notes


Teaching

Graduate Student Instructor, UC Berkeley
  • CS 294-165: Sketching Algorithms (Fall 20)
  • CS 170: Efficient Algorithms and Intractable Problems (Spring 20)
Undergraudate Teaching Assistant, Duke University

Contact

z0@berkeley.edu