Portrait
Shao Chenxi
M.S. Student, East China Normal University
School of Computer Science and Technology
Visual Perception + X (VPX) Group
About Me

I am a Master's student at the School of Computer Science and Technology, East China Normal University (ECNU), supervised by Prof. Changbo Wang and Assoc. Prof. Yang Li. My research interests include video content understanding, multimodal large models, vision foundation models, and computer vision. Currently, my work focuses on streaming video understanding, memory management mechanisms in vision-language models, and multi-task training for vision-language models.

Before joining ECNU, I received my B.S. in Computer Science and Technology from Huazhong Agricultural University. During my undergraduate years, I was an active competitive programmer and won awards in multiple national programming contests. In addition, I have full-stack AI system development experience spanning from prototype to deployment, applying computer vision to real-world scenarios such as psychological assessment and industrial systems.

If you are interested in my work, feel free to reach out via email!

Education
  • East China Normal University
    East China Normal University
    School of Computer Science and Technology
    M.S. in Computer Science and Technology
    Sep. 2025 - Present
  • Huazhong Agricultural University
    Huazhong Agricultural University
    B.S. in Computer Science and Technology
    Sep. 2021 - Jun. 2025
Experience
  • MIGU Video Co., Ltd.
    China Mobile Migu / Migu Video
    AI Algorithm Engineer Intern
    Apr. 2026 - Jul. 2026
    Developed AI capabilities for football live streams, spanning video captioning, streaming commentary, player tracking, and close-up identity recognition.
Honors & Awards
National Scholarship for Undergraduate Students
2022
CCF Elite Collegiate Award
2024
49th International Collegiate Programming Contest Asia Shenyang Regional Contest Bronze Medal
Nov. 2024
China Collegiate Programming Contest National Invitational Contest (Zhengzhou) Silver Award
May 2024
China Collegiate Programming Contest National Invitational Contest (Shandong) Silver Award
May 2024
15th Lanqiao Cup National Final (C/C++ Group A) Second Prize
Jun. 2024
Group Programming Ladder Tournament National Final Second Prize
2024
Selected Publications (view all )
StreamSoccer: Event-Driven Memory for Streaming Soccer Commentary
StreamSoccer: Event-Driven Memory for Streaming Soccer Commentary

Chenxi Shao, Bozhong Wang, Jiaxin Huang, Zhao Liu, Sunwei Zhu, Tianxin Hang, Gaoqi He, Yang Li#, Changbo Wang# (# corresponding author)

arXiv preprint Aug. 2026

We present StreamSoccer, an event-driven system for causal streaming soccer commentary. It maintains fixed-budget event memory, preserves completed events as local context, and consolidates them into retrievable historical records for current-event, recent-window, and historical-memory commentary.

StreamSoccer: Event-Driven Memory for Streaming Soccer Commentary

Chenxi Shao, Bozhong Wang, Jiaxin Huang, Zhao Liu, Sunwei Zhu, Tianxin Hang, Gaoqi He, Yang Li#, Changbo Wang# (# corresponding author)

arXiv preprint Aug. 2026

We present StreamSoccer, an event-driven system for causal streaming soccer commentary. It maintains fixed-budget event memory, preserves completed events as local context, and consolidates them into retrievable historical records for current-event, recent-window, and historical-memory commentary.

Mind the Context: Attention-Guided Weak-to-Strong Consistency for Enhanced Semi-Supervised Medical Image Segmentation
Mind the Context: Attention-Guided Weak-to-Strong Consistency for Enhanced Semi-Supervised Medical Image Segmentation

Yuxuan Cheng, Chenxi Shao, Jie Ma, Yunfei Xie, Guoliang Li

arXiv preprint Oct. 2024

Medical image segmentation relies on high-quality annotated data that is often difficult and costly to obtain. To address this, we propose an Attention-Guided weak-to-Strong Consistency Match (AIGCMatch) framework that incorporates attention-guided perturbation strategies at both the image level and feature level to achieve weak-to-strong consistency regularization. This approach preserves the structural information of medical images while enhancing the model's ability to process complex semantic information. Extensive experiments on the ACDC and ISIC-2017 datasets validate the effectiveness of our method, achieving a 90.4% Dice score on ACDC in the 7-case scenario and significantly outperforming baselines on ISIC-2017.

Mind the Context: Attention-Guided Weak-to-Strong Consistency for Enhanced Semi-Supervised Medical Image Segmentation

Yuxuan Cheng, Chenxi Shao, Jie Ma, Yunfei Xie, Guoliang Li

arXiv preprint Oct. 2024

Medical image segmentation relies on high-quality annotated data that is often difficult and costly to obtain. To address this, we propose an Attention-Guided weak-to-Strong Consistency Match (AIGCMatch) framework that incorporates attention-guided perturbation strategies at both the image level and feature level to achieve weak-to-strong consistency regularization. This approach preserves the structural information of medical images while enhancing the model's ability to process complex semantic information. Extensive experiments on the ACDC and ISIC-2017 datasets validate the effectiveness of our method, achieving a 90.4% Dice score on ACDC in the 7-case scenario and significantly outperforming baselines on ISIC-2017.

All publications
Selected Projects
MentorForge — Research Mentor Distillation for AI Agents
Jul. 2026
Open-source Project (13 ★)
An open-source Codex Skill that distills a scholar's publications, homepage, and academic trajectory into a runnable research operating system for AI agents. Rather than mimicking how a mentor talks, it captures how a mentor makes research judgments — what problems are worth doing, what experiments are convincing, and the methodology across research directions. Multi-source publication collection, full-text extraction, cross-paper synthesis, with evidence-grounded verification.
Intelligent Sandplay for Psychological Assessment
Sep. 2025 - Dec. 2025
Video Analysis and System Development
Worked on an AI-assisted sandplay analysis system involving object detection, multi-object tracking, and visualization for long-sequence scene understanding.
Multi-type Meter Visual Recognition and Automated Billing System
Mar. 2025 - Dec. 2025
Algorithm and Full-stack Development
Built components for an industrial meter-reading workflow, combining visual localization and OCR with backend and mini-program interfaces for task management and billing reports.
Research on Pulmonary Nodule Detection and Segmentation based on CT Images
Hubei Provincial Undergraduate Training Program
Research Participant
Worked on CT image preprocessing and segmentation experiments based on NestedU-Net and UNet-style models using the LIDC-IDRI dataset.
News
2026
Released StreamSoccer, an event-driven system for causal streaming soccer commentary, together with its open-source code and project page.
Aug. 2026
Completed the AI Algorithm Engineer Internship at MIGU Video Co., Ltd., contributing to football video understanding and intelligent commentary generation.
Aug. 2026
Open-sourced MentorForge: a Skill that distills a scholar's publications and academic trajectory into a runnable research mentor for AI agents, guiding research judgment under evidence constraints.
Jul. 2026
Started an internship as an AI Algorithm Engineer Intern at MIGU Video Co. Ltd.
Apr. 2026
2025
Joined the VPX Group at East China Normal University as a Master's student.
Sep. 2025
Graduated from Huazhong Agricultural University with a B.S. in Computer Science and Technology.
Jun. 2025
2024
Won the Bronze Medal in the 49th ICPC Asia Shenyang Regional Contest.
Nov. 2024
Selected for the CCF Elite Collegiate Award.
Aug. 2024
Won Second Prize in the 15th Lanqiao Cup National Final (C/C++ Group A).
Jun. 2024
Won Silver Awards in the CCPC National Invitational Contests.
May 2024