Back to Home

Yubo Wang

📧 bobwang@toki.waseda.jp 📍 Kanagawa, Japan 🎂 July 23, 1998
LinkedIn • GitHub • HuggingFace

Professional Summary

Ph.D. candidate specializing in Computer Vision and Multimodal AI. Research expertise in vision-language models, generative models for high-resolution image synthesis, and semantic segmentation under domain shift challenges. Proficient in LLM/VLM development with hands-on experience in Model Context Protocol (MCP) integration, vLLM deployment, and LoRA-based supervised fine-tuning. Strong track record of translating research into industrial solutions for OPPO, Bosch, and NTT.

Education

Waseda University
Oct. 2022 – Present
Ph.D. in Graduate School of Creative Science and Engineering
Tokyo, Japan
Advisors: Prof. Hiroyuki Ishii, Prof. Sugano Shigeki
Waseda University
Oct. 2020 – Sept. 2022
M.Eng. in Graduate School of Creative Science and Engineering (GPA: 3.73/4.00)
Tokyo, Japan
Advisor: Prof. Jun Ohya
Shandong University (985 Project)
Sept. 2016 – June 2020
B.Eng. in Control Science and Engineering (GPA: 4.01/5.00)
Shandong, China
Advisor: Prof. Guoliang Liu

Technical Skills

Languages & Frameworks:
Python, PyTorch, LangChain, LangGraph, vLLM, LLaMA-Factory
Cloud & Infrastructure:
Linux, AWS (EC2/S3), Slurm, ABCI, Jetson AGX Orin
Development & Tools:
Git, Model Context Protocol (MCP), HuggingFace, LoRA/QLoRA

Research Areas

Generative Models (Diffusion/Flow Matching), Multimodal Learning (Qwen3-VL, Deepseek-OCR), LLM Agents & Tool Use, Semantic Segmentation, Semi-Supervised Learning, Multi-Object Tracking (MOT)

Experience

OPPO, Japan Research Center
Sept. 2025 – Present
Research Engineer Internship | Yokohama, Japan
  • Developing next-generation camera algorithms for mobile platforms
  • Optimized Multi-Object Tracking (MOT) pipelines by integrating ReID embeddings into Joint-Detection architectures
University of Maryland (UMIACS)
Sept. 2024 – Feb. 2025
Research Scholar | Maryland, U.S.A
Advisor: Prof. Abhinav Shrivastava
Research: High-Resolution Image Synthesis & Representation Learning
  • Trained an Img-to-Img Diffusion Transformer (DiT) from scratch for 1024×1024 aerial disaster scene generation
  • Orchestrated distributed training on a 4×A6000 GPU cluster
Bosch
Nov. 2023 – Jan. 2024
Research Engineer Internship | Yokohama, Japan
  • ADAS R&D: Addressed Domain Shift in autonomous driving using Generative Data Augmentation
  • Enhanced lane detection robustness with Consistency Regularization
Incubit Inc.
Dec. 2022 – May 2023
Research Engineer Internship | Tokyo, Japan
  • Diffusion model-based medical brain MRI recognition, focusing on tumor segmentation
Waseda–NTT–JAXA Joint Research
Feb. 2021 – May 2023
Computer Vision Researcher | Tokyo, Japan
Mentor: Dr. Zhao Wang
  • Built AI systems for disaster response using JAXA satellite imagery
  • Proposed Multi-Scale Attention Cascade model achieving SOTA in aerial segmentation
  • Developed context-enhanced models for traffic jam detection
  • Published 3 Japan patents

Honors and Awards

Publications

For a complete list of publications, please visit the Publications page.