Huiwen Wu

Associate Researcher · Beihang University, Hangzhou International Innovation Institute

I build trustworthy foundation models through the lens of numerical analysis, working at the intersection of differential privacy, optimization theory, and AI for science.

My research combines PhD training in randomized solvers (UC Irvine, Mathematics) with industrial-scale deployment experience at Ant Group, and applies it to privacy-preserving large language models and scientific foundation models. The through-line is that the tools of classical numerical analysis — preconditioning, multigrid, randomized sketching — turn out to be exactly what modern machine learning needs when it has to be both private and large.

I have published 10+ first- or corresponding-author papers at top-tier venues including AAAI, WWW and IJCAI, hold authorized US patents in privacy-preserving model training, and serve as PI on a Zhejiang Lab talent grant. I am a core contributor to National Key R&D Programs, including AI-powered galaxy simulation for the Chinese Space Station Telescope.

Research interests

Selected work

Multigrid–ViT Fusion for Seismic Foundation Models

AAAI'26 · CCF-A
96.4% reduction in MAE (0.1886 → 0.0068) against standard ViT baselines on foundation-model pretraining for seismogram reconstruction.

Communication-Efficient and Differentially Private Federated Fine-tuning for LLMs

AAAI'25, ECAI'25 · CCF-A/B
Near-non-private accuracy on C-Eval (+1.2%) and MMLU (+0.8%) at ε = 0.25 — closing a utility gap where prior DP-LLM approaches (e.g. DP-FedLoRA) lose more than 50% at the same ε.

Differentially Private Knowledge Transfer for Recommendation

WWW'22 · CCF-A
+8.43% accuracy over the private state of the art (PriCDR-SYM) on Amazon cross-domain benchmarks at ε = 0.5.

Read more about these projects →

Teaching

Matrix Theory and Applications I

M571011S03 Graduate 16 weeks · 32 contact hours 2026–
A graduate course on matrix theory and matrix computations for students in artificial intelligence and related disciplines, built on Golub & Van Loan's Matrix Computations. Course website →

All teaching →

Background

Now
Associate Researcher, Beihang University, Hangzhou International Innovation Institute (2026–)
Before
Senior Researcher, Zhejiang Laboratory (2022–26) · Senior Algorithm Engineer, Ant Group (2020–22)
PhD
Mathematics, University of California, Irvine (2019) — randomized fast solvers, advised by Prof. Long Chen
Undergraduate
Mathematics, Sichuan University (2013), with Honors

Full CV →