CV

Education

Fudan University, M.Eng. in Electronic Information, 2027 - 2030 (expected)

Nankai University, Dual Bachelor's Degrees in Engineering and Law, 2023 - 2027 (expected)

College of Cryptology and Cyber Science, Information Security and Law dual-degree program
Research advisor: Prof. Yu Zhou

Scene Text Tracking, Prof. Yu Zhou's Group, Nankai University, Aug 2025 - Present

To address deformation, blur, and structural-detail loss, proposed SymTrack, the first detection-free framework for scene text tracking, and built a dedicated benchmark from three video text spotting datasets.
Independently completed the literature review, feasibility study, PyTorch implementation, ablation studies, reproduction of more than ten recent state-of-the-art trackers, full evaluation, and paper writing.
Achieved state-of-the-art results on all three benchmarks. The work was accepted by ICML 2026 with me as sole first author.

Video Text Editing, Prof. Yu Zhou's Group, Nankai University, Jan 2026 - Present

To improve glyph rendering and reduce background distortion, developed the first training-free video text editing framework based on Wan 2.2, using frequency-guided glyph injection and mask-trajectory preservation. It outperformed VACE, FlowEdit, and FlowAlign in overall quality.
Led the construction of a benchmark containing real and synthetic data, reproduced multiple leading methods, and completed the implementation, experiments, and manuscript. The work is under review at AAAI 2027.

Scene Text Image Super-Resolution, Prof. Yu Zhou's Group, Nankai University, Dec 2025 - Jun 2026

To mitigate visual hallucinations caused by overreliance on semantic priors, helped develop a hallucination-resistant dual-branch diffusion framework with a three-source evidence gating module for adaptive structural and semantic fusion. The method achieved state-of-the-art results on CTR and Real-CE under ×2 and ×4 settings.
Contributed to the PyTorch implementation, ablation studies, and manuscript. The second-author paper was accepted by ACM Multimedia 2026.

Document Intelligence, CCF Student Pilot Program, Prof. Ming-Ming Cheng's Group, Nankai University, Sep 2025 - Feb 2026

Studied handwritten formula recognition and table parsing, reproduced recent state-of-the-art models, and ported TAMER (AAAI 2025) from PyTorch to Jittor with aligned accuracy.
Completed the program and passed the final presentation.

Unified Visual Understanding and Generation, Prof. Wenguan Wang's Group, Zhejiang University, May 2026 - Present

Investigating a unified-representation framework that addresses conflicts between discriminative visual understanding and generative modeling.
Unifying diverse conditional inputs and outputs so that a single model can perform conditional generation alongside visual tasks such as object detection, semantic segmentation, and depth estimation.

Beyond Detection: A Structure-Aware Framework for Scene Text Tracking. Chenmin Yu, Liu Yu, Daiqing Wu, Gengluo Li, Zeyu Chen, and Yu Zhou. ICML 2026. Sole first author. Paper · Code
Structure Leads, Semantics Assist: Hallucination-Resistant Diffusion for Scene Text Image Super-Resolution. Zeyu Chen, Chenmin Yu, Fangmin Zhao, Yichao Liu, and Yu Zhou. ACM Multimedia 2026. Second author.

First Prize, Tianjin Division, China Undergraduate Mathematical Contest in Modeling (2025)
Meritorious Winner, Mathematical Contest in Modeling (2025)

Deputy Director, Publicity Department, Youth League Committee of the College of Computer Science and College of Cryptology and Cyber Science
Outstanding Officer, Extracurricular Activities Guidance Center of the Nankai University Youth League Committee

Research: Literature review, paper reproduction, experimental design, academic writing
Programming and tools: Python, PyTorch, Linux, Jittor, LaTeX, Git
English: CET-6, 559