Yuhao Su

Yuhao Su is a PhD candidate in Computer Science at Khoury College of Computer Sciences at Northeastern University, advised by Prof. Ehsan Elhamifar. His research focuses on multimodal LLMs, video understanding, and agentic and interactive AI.

He was a Research Intern at Samsung Research America (2026) and United Imaging Intelligence (2025). Before Northeastern, he earned his B.A. in Mathematics and Computer Science from the University of Minnesota.

He is seeking full-time Applied/Research Scientist, MLE, and SWE roles starting 2027.

News

Jun 2026	MedVidBench serves as the benchmark for the MedVidU Challenge at ECCV 2026.
May 2026	Started research internship at Samsung Research America, Mountain View.
Apr 2026	Passed thesis proposal.
Mar 2026	MedGRPO accepted to CVPR 2026.
Oct 2025	RegionAligner accepted to WACV 2026.

Selected Works

MedGRPO: Multi-Task Reinforcement Learning for Heterogeneous Medical Video Understanding

Yuhao Su, Anwesa Choudhuri, Zhongpai Gao, Benjamin Planche, Van Nguyen Nguyen, Meng Zheng, Yuhan Shen, Arun Innanje, Terrence Chen, Ehsan Elhamifar, Ziyan Wu

A 531K-pair medical video benchmark (MedVidBench) covering 8 tasks, plus a multi-task RL post-training framework whose cross-dataset reward normalization keeps training stable when heterogeneous tasks are mixed. MedVidBench is the benchmark of the ECCV 2026 MedVidU Challenge.

CVPR 2026

[Paper] [Project] [Code]

RegionAligner: Bridging Ego-Exo Views for Object Correspondence via Unified Text-Visual Learning

Yuhao Su, Ehsan Elhamifar

Matches the same object between first-person and third-person views: a VLM names the task-relevant objects to narrow the search region, and multi-level supervision aligns regions across views. Works in both supervised and unsupervised settings on Ego-Exo4D.

WACV 2026

[Paper] [Code]

Two-Stage Active Learning for Efficient Temporal Action Segmentation

Yuhao Su, Ehsan Elhamifar

Cuts annotation cost for temporal action segmentation by first choosing the most diverse videos, then the frames that best summarize each one. Reaches 95% of fully supervised performance with only 0.35% of frames labeled.

ECCV 2024

[Paper] [Code]

Under Review

Ongoing research on agentic video LLMs, interactive video understanding, and multimodal generation.

Academic Service

Reviewer: CVPR, ICCV, ECCV, NeurIPS, ICLR, TPAMI, WACV, ICASSP