Wenyuan Zhao

ROOM 326

WEB Building

College Station, TX 77840

wyzhao[at]tamu[DOT]edu

Hi, I am Wenyuan(Warren), a third year Ph.D. student in the Electrical and Computer Engineering Department at Texas A&M University, advised by Professor Chao Tian. Currently, I am working on the general area of Machine Learning and Information Theory where new algorithms are proposed dealing with diverse settings.

Before joining TAMU, I obtained my M.S. degree in Communication Theory & Systems at UC San Diego, where I worked on Dynamic mmWave Networking with Professor Xinyu Zhang. Before that, I obtained my B.E. degree in Information Engineering at Southeast University in 2021.

My research interests lie in the general area of machine learning and information theory. Particularly, I am working on uncertainty quantification, multimodal learning, and their applications in large language models and generative foundation models.

News

Apr 30, 2026	Three first authored/co-authored papers (GPan-LoRA, SIKA-GP, Trust3R) are accepted by ICML 2026!
Nov 14, 2025	ECE Poster Event Winner @ Texas A&M University! (Two years in a row!)
Oct 11, 2025	NeurIPS 2025 Scholar Award!
Sep 18, 2025	“Partial Information Decomposition via Normalizing Flows in Latent Gaussian Distributions” is accepted by NeurIPS 2025!
Jun 18, 2025	W-PIR is accepted by IEEE Transactions on Information Theory!
Apr 28, 2025	ISIT 2025 Travel Grant Award!
Jan 21, 2025	“From deep additive kernel learning to last-layer Bayesian neural networks via induced prior approximation” is accepted by AISTATS 2025!
Aug 15, 2023	Pursuing Ph.D. degree at Texas A&M University with Prof. Chao Tian!
Nov 10, 2019	First-prize winner at 2019 Mathematical Contest in Modeling!

Selected Publications

ICML

SIKA-GP: Accelerating Gaussian Process Inference with Sparse Inducing Kernel Approximations for Bayesian Deep Learning

Wenyuan Zhao, Rui Tuo, and Chao Tian

The Forty-Third International Conference on Machine Learning, 2026

Bib PDF Code

@article{zhao2026sika,
  title = {SIKA-GP: Accelerating Gaussian Process Inference with Sparse Inducing Kernel Approximations for Bayesian Deep Learning},
  author = {Zhao, Wenyuan and Tuo, Rui and Tian, Chao},
  journal = {The Forty-Third International Conference on Machine Learning},
  year = {2026},
}

ICML

GPan-LoRA: Gaussian Process Amortized Networks for Bayesian Low-Rank Adaptation in Large Language Models

Weifeng Zhang^*, Wenyuan Zhao^*, Amir Hossein Rahmati, Yucheng Wang, Zhiyuan Wang, Chao Tian, and Xiaoning Qian

The Forty-Third International Conference on Machine Learning, 2026

Bib PDF Code

@article{zhang2026gpan,
  title = {GPan-LoRA: Gaussian Process Amortized Networks for Bayesian Low-Rank Adaptation in Large Language Models},
  author = {Zhang, Weifeng and Zhao, Wenyuan and Rahmati, Amir Hossein and Wang, Yucheng and Wang, Zhiyuan and Tian, Chao and Qian, Xiaoning},
  journal = {The Forty-Third International Conference on Machine Learning},
  year = {2026},
}

ICML

Trust3R: Unifying Feed-Forward Pointmap Prediction and Evidential Learning for Trust-Aware 3D Reconstruction

Zihao Zhu^*, Wenyuan Zhao^*, Nuo Chen, Chao Tian, and Zhiwen Fan

The Forty-Third International Conference on Machine Learning, 2026

Bib PDF Code

@article{zhu2026trust3r,
  title = {Trust3R: Unifying Feed-Forward Pointmap Prediction and Evidential Learning for Trust-Aware 3D Reconstruction},
  author = {Zhu, Zihao and Zhao, Wenyuan and Chen, Nuo and Tian, Chao and Fan, Zhiwen},
  journal = {The Forty-Third International Conference on Machine Learning},
  year = {2026},
}

NeurIPS

Partial Information Decomposition via Normalizing Flows in Latent Gaussian Distributions

Wenyuan Zhao, Adithya Balachandran, Chao Tian, and Paul Pu Liang

The Thirty-ninth Annual Conference on Neural Information Processing Systems, 2025

Bib PDF Code

@article{zhao2025partial,
  title = {Partial Information Decomposition via Normalizing Flows in Latent Gaussian Distributions},
  author = {Zhao, Wenyuan and Balachandran, Adithya and Tian, Chao and Liang, Paul Pu},
  journal = {The Thirty-ninth Annual Conference on Neural Information Processing Systems},
  year = {2025},
}