Yipeng Zhang

Senior AI Engineer building agentic, multimodal, and clinical AI systems

I am currently a Senior AI Engineer at Zoom, working on agentic workflows, LLM-based user simulation, and large-scale evaluation of automated systems. Previously, I was an Applied Researcher 2 at eBay Search Science, leading end-to-end development of multimodal product understanding and agentic search.

I received my Ph.D. in Electrical and Computer Engineering from University of California, Los Angeles under the supervision of Prof. Vwani Roychowdhury. Before that, I earned my M.S. from University of California, San Diego and my B.S. from University of Wisconsin-Madison.

I drive end-to-end ownership of production AI systems, from problem formulation and model development to deployment, evaluation, and continuous iteration at scale. My work focuses on building reliable LLM and multimodal pipelines with rigorous evaluation and measurable product impact in cross-functional environments. In parallel, I develop interpretable EEG/iEEG models for epilepsy in clinical settings, and I am the main developer of PyHFO, an open-source platform for clinically grounded high-frequency oscillation analysis.

Across my research and applied work, I have authored 10+ first/co-first author publications and filed multiple U.S. patents.

profile photo
Agentic AI Multimodal Search LLM Eval Medical AI

Publications

Selected work across multimodal systems, search, and clinically grounded EEG/iEEG research.

Jump to year
2026
Selected
ICLR 2026
Omni-iEEG icon
Omni-iEEG: A Large-Scale, Comprehensive iEEG Dataset and Benchmark for Epilepsy Research
Chenda Duan*, Yipeng Zhang*, Sotaro Kanai, Yuanyi Ding, et al.
* Co-first authors
International Conference on Learning Representations (ICLR), 2026  
Paper / Code / Website

WSDM 2026
REFLEX icon
REFLEX: Reinforcement Feedback Learning with Large Language Models for E-commerce Query Expansion
Yipeng Zhang, Bowen Liu, Xiaoshuang Zhang, et al.
ACM International Conference on Web Search and Data Mining (WSDM), 2026  
Project Page / Paper

ICME 2026
VoidFormer figure
VoidFormer: Mitigating Attention Distortion for Multimodal Learning Under Missing Modality
Jinglong Yang, Penghui Zhou, Yipeng Zhang, et al.
IEEE International Conference on Multimedia and Expo (ICME), 2026  
Coming Soon

WWW 2026
SEER icon
SEER: Set Encoding for Efficient Representation in Large-Scale E-commerce.
Yining Lin, Yuming Shen, Yipeng Zhang, Canran Xu
The Web Conference (WWW), 2026  
Project Page / Paper

2025
Selected
ICDM MMSR 2025
Bridging modality gaps figure
Bridging Modality Gaps in e-Commerce Products via Vision-Language Alignment
Yipeng Zhang, Hongju Yu, Aritra Mandal, Canran Xu, et al.
ICDM MMSR, 2025  

MICCAI 2025
SS2DL icon
Self-Supervised Distillation of Legacy Rule-Based Methods for Enhanced EEG-Based Decision-Making
Yipeng Zhang, Yuanyi Ding, Chenda Duan, Atsuro Daida, et al.
International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), 2025  
Paper

2024
Selected
JNE 2024
PyHFO icon
PyHFO: lightweight deep learning-powered end-to-end high-frequency oscillations analysis application
Yipeng Zhang, Lawrence Liu, Yuanyi Ding, et al.
Journal of Neural Engineering (JNE), 2024  
Paper

2023
Selected
Sci Rep 2023
Scientific Reports icon
Decoding of human identity by computer vision and neuronal vision
Yipeng Zhang, Zahra M. Aghajan, Matias Ison, et al.
Scientific Reports, 2023  
Paper

2022
Selected
ACM MM 2022
ODMO icon
Action-conditioned On-demand Motion Generation
Qiujing Lu*, Yipeng Zhang*, Mingjian Lu, Vwani Roychowdhury
* Co-first authors
ACM International Conference on Multimedia (ACM MM), 2022  
Paper / Project Page / Code

2021
Selected
Brain Comms 2021
eHFO icon
Refining epileptogenic high-frequency oscillations using deep learning: a reverse engineering approach
Yipeng Zhang, Qiujing Lu, Tonmoy Monsoor, et al.
Brain Communications, 2021  
Paper / Code

Visitor Map