Song Wen

Song Wen

I am currently a Research Scientist at Meta. My research spans Generative AI, Multimodal Large Language Models, Neural ODEs, and Computer Vision, with a focus on model development, training, and data pipelines.

I received my Ph.D. in Computer Science from Rutgers University, where I was advised by Prof. Dimitris N. Metaxas. Prior to that, I obtained my M.Phil. degree from the Hong Kong University of Science and Technology (HKUST) and my B.Eng. degree from Huazhong University of Science and Technology (HUST).

Research Interests

Generative AI

Diffusion models, flow matching, rectified flows, controllable generation, image synthesis and editing

Multimodal LLMs

Multimodal reasoning, instruction tuning, omni-modal unified models

Neural ODEs

Trajectory forecasting, multi-agent interaction modeling, dynamical systems

Computer Vision

Efficient neural architectures, image and video restoration, image segmentation

Industry Experience

MetaResearch Scientist
AdobeResearch Intern
Sony AIResearch Intern
Shanghai AI LabResearch Intern
TencentResearch Intern

Selected Publications

Full list on Google Scholar.

WACV 2026

DICE: Discrete Inversion Enabling Controllable Editing for Masked Generative Models

X He, Q Dao, L Han, S Wen, M Bai, D Liu, H Zhang, F Juefei-Xu, C Tan, et al.

Meas. Sci. Tech.

Roadmap: Integrating Artificial Intelligence in Structural Health Monitoring Systems

S Laflamme, E Blasch, F Ubertini, Z Liu, ..., S Wen, et al.

ICCV 2025
ICLR 2025

Implicit In-context Learning

Z Li, Z Xu, L Han, Y Gao, S Wen, D Liu, H Wang, DN Metaxas

WACV 2025

SODA: Spectral Orthogonal Decomposition Adaptation for Diffusion Models

X Zhang, S Wen, L Han, F Juefei-Xu, A Srivastava, J Huang, V Pavlovic, et al.

CVPR-W 2025
WACV 2024

ProxEdit: Improving Tuning-Free Real Image Editing with Proximal Guidance

L Han, S Wen, Q Chen, Z Zhang, K Song, M Ren, R Gao, A Stathopoulos, et al.

WACV 2024

Second-Order Graph ODEs for Multi-Agent Trajectory Forecasting

S Wen, H Wang, D Liu, Q Zhangli, D Metaxas

J. Urban Mobility

The Traffic Calming Effect of Delineated Bicycle Lanes

H Younes, C Andrews, RB Noland, J Xia, S Wen, W Zhang, D Metaxas, et al.

CVPR 2023

Run, Don't Walk: Chasing Higher FLOPS for Faster Neural Networks

J Chen, S Kao, H He, W Zhuo, S Wen, CH Lee, SHG Chan

arXiv 2023

ImageBind-LLM: Multi-modality Instruction Tuning

J Han, R Zhang, W Shao, P Gao, P Xu, H Xiao, K Zhang, C Liu, S Wen, et al.

arXiv 2023
ECCV 2022
MICCAI 2022

Transfusion: Multi-view Divergent Fusion for Medical Image Segmentation with Transformers

D Liu, Y Gao, Q Zhangli, L Han, X He, Z Xia, S Wen, Q Chang, Z Yan, et al.

MICCAI 2022

Region Proposal Rectification towards Robust Instance Segmentation of Biological Images

Q Zhangli, J Yi, D Liu, X He, Z Xia, Q Chang, L Han, Y Gao, S Wen, et al.

DDDAS 2022
AAAI 2021
US Patent
arXiv 2021
ICCV-W 2019

Professional Service

Reviewer / Program Committee Member

NeurIPS2022, 2023, 2024, 2025
ICML2024, 2025
ICLR2024, 2025
CVPR2024, 2025, 2026
ICCV2025
ECCV2022, 2024, 2026
AAAI2025, 2026
KDD2022, 2023, 2024
WACV2026
ECMLPKDD2023, 2024
SDM2024