Xiaomeng Yang 杨晓萌

Ph.D. Candidate in Computer Engineering

Northeastern University · Boston, MA

Xiaomeng Yang

I am a Ph.D. candidate in Computer Engineering at Northeastern University, co-advised by Prof. Yanzhi Wang and Prof. Xuan Zhang. My research interests span video generation, world models, efficient generative models, multimodal learning, and visual understanding. Recently, I have been working on world action models for robot manipulation and on efficient diffusion models for video generation.

Previously, I earned my master's degree at the University of Chinese Academy of Sciences, where I worked with Prof. Yu Zhou on visual text analysis and understanding. I hold dual bachelor's degrees in Computer Engineering from the University of Illinois Urbana-Champaign and Zhejiang University.

I'm always glad to connect about research ideas or potential collaborations — feel free to reach out by email.

News

Selected Publications

Event-Aligned Visual Action Reasoning for World Action Models

Xiaomeng Yang, Yushu Wu, Yi Gao, Yuhao Lei, Xuan Zhang, Pu Zhao, Yanzhi Wang

arXiv preprint, 2026

An event-aligned visual action reasoning framework that organizes visual-action prediction around interaction events, with an execution validity head that identifies the valid portion of each predicted action sequence.

Paper Project Code

Prompt2Effect: Training-Free Image-to-Video Model Specialization via LoRA Generation

Xiaomeng Yang, Yanyu Li, Gordon Guocheng Qian, Ivan Skorokhodov, Viacheslav Ivanov, Avalon Vinella, Xuan Zhang, Yanzhi Wang, Sergey Tulyakov, Anil Kag

European Conference on Computer Vision (ECCV), 2026

A training-free framework that specializes image-to-video diffusion models via generated LoRAs for controllable video effects.

Paper Project

ALTER: All-in-one Layer Pruning and Temporal Expert Routing for Efficient Diffusion Generation

Xiaomeng Yang*, Lei Lu*, Qihui Fan, Changdi Yang, Juyi Lin, Yanzhi Wang, Xuan Zhang, Shangqian Gao (* equal contribution)

Conference on Neural Information Processing Systems (NeurIPS), 2025

Combines all-in-one layer pruning with temporal expert routing to accelerate diffusion-based generation.

Paper Code

IPAD: Iterative, Parallel, and Diffusion-based Network for Scene Text Recognition

Xiaomeng Yang, Zhi Qiao, Yu Zhou

International Journal of Computer Vision (IJCV), 2025

An iterative, parallel, and diffusion-based network for accurate and efficient scene text recognition.

Paper Code

Experience

Research Intern, Creative Vision, Snap Inc.

Sep 2025 – May 2026 · Santa Monica, CA

Worked on efficient image-to-video effects generation.

Research Intern, Tomorrow Advancing Life (TAL)

Nov 2022 – Jun 2024 · Beijing, China

Worked on scene text recognition.

Education

Northeastern University
Ph.D. in Computer Engineering · 2024 – Present
University of Chinese Academy of Sciences
M.S., advised by Prof. Yu Zhou · 2021 – 2024
University of Illinois Urbana-Champaign
B.S. in Computer Engineering (dual-degree) · 2017 – 2021
Zhejiang University
B.S. in Computer Engineering (dual-degree) · 2017 – 2021