Tsz-To Wong Aiden

M.S. Student at NYCU & NSTC Research Assistant.

I turn open-ended research questions into things people can test, use, and learn from.

Worklog

What I've actually built, organised by the skills it took to build it.

Skills
Languages
Projects

LocalSubs: On-Device Real-Time Subtitle Translation

A fully local English-to-Taiwan-Traditional-Chinese subtitle translator, built around a workload-optimized 0.6B model and a Chrome-to-native inference path.

LLMMachine LearningGoML Systems / Deployment

AnimateQR: Functionality-Preserving Animated QR Code Generation

Open-source system for generating animated QR code videos that remain machine-scannable — built around a staged pipeline that decouples video motion synthesis from QR structural preservation, with ArtCoder-based multi-objective refinement.

Computer VisionMachine LearningPython

P Language Compiler from Scratch

End-to-end compiler for the P language, covering lexical analysis through RISC-V assembly emission — built to understand every stage of the compilation pipeline by implementing it.

C++

Contrastive Guidance for Diffusion Restoration

Training-free inference-time semantic guidance for instruction-following diffusion image restoration, using CLIP-based contrastive signals to improve prompt alignment and perceptual quality.

Computer VisionMachine LearningPython

Education

From computer science fundamentals to applied AI research.

2024 – Present

M.S. Computer Science

National Yang Ming Chiao Tung University (NYCU)·Hsinchu

Vision-language models, agentic system applications, real-time embedded systems, and parallel programming

Vision-Language ModelsAgentic SystemsReal-Time Embedded SystemsParallel Programming
2020 – 2024

B.S. Computer Science and Engineering

National Sun Yat-sen University (NSYSU)·Kaohsiung

Focused on computer science fundamentals

Blog

Notes on model performance, inference systems, and the engineering behind research prototypes.

2026-08-01

Why a 0.6B Model Still Felt Slow

localsubsperformance 2025-10-20

Building a Streaming Pipeline for BLIP-2–Flan-T5

multimodalstreaming
View all →

Publications & Preprints

2026

From Global Alignment to Local Grounding: Zero-Shot Chinese Character Recognition with Radical Verification

Yu-Heng Shih , Bing-Chen Wu , Tsz-To Wong , Ting-En Yen , Hong-Han Shuai , Bin-Hua Hsieh , ChienAn Chen , Yi-Ren Yeh , Ching-Chun Huang
ACCV 2026 · Oral
2026

Workload-Driven Optimization for On-Device Real-Time Subtitle Translation

Tsz-To Wong
arXiv:2607.09957 [cs.CL]
2025

COACH: Collaborative Agents for Contextual Highlighting — A Multi-Agent Framework for Sports Video Analysis

Tsz-To Wong , Ching-Chun Huang , Hong-Han Shuai
AAAI 2026 Workshop LaMAS

Contact

Email is the best way to reach me.

GitHub@aiden1020LinkedInin/tsz-to-wong-a28502411ScholarTsz-To Wong