STEER framework
AI4R experiment framework for auto-research / autonomous research under governance (frozen task, scorer, evidence).
Yulai Xie · 谢雨来
PhD, Researcher
China Telecom Research Institute
I am a Researcher at China Telecom Research Institute. I received my Ph.D. in Engineering from Hokkaido University, Japan (CSC Scholarship), and worked at Hitachi China Research Laboratory from 2014 to 2025, most recently as a Chief Researcher on CV / AI / AIGC industrial applications and GenAI / LLM exploration. My research interests span machine learning and generative AI, applied to industrial intelligence, logistics, crowd simulation, and optics.
News
2026
2025
- Dec Media Paper accepted by Knowledge-Based Systems
- Dec Paper Unconstrained Vision-based Measurement — ICCMS 2025
- Nov Paper Structured guided diffusion models — Knowledge-Based Systems
- Sep Paper Query-based neural network for multimode-fiber dynamics — ESWA
- Aug Career Joined China Telecom Research Institute
- Apr Media ChinaDaily Media Report — MCVR2024
- Mar Service Technical committee (ICCMS2025)
2024
- Dec Talk Invited talk (MCVR2024)
- Jul Paper Top Downloaded Article in IET Computer Vision
- Jun Media Digital simulation empowers industrial AI solutions
- Mar Media Hitachi Review
- Mar Media Hitachi Review (日本語)
- Jan Service Technical committee (ICCMS2024)
2023
- Nov Media Medical Exhibition
- Jul Media AI Industrial Application
- Jan Service Chairman (ICCSM2023)
2022
- Aug Talk Invited talk report
- Jul Talk Invited talk (ICCCV2022)
2020
- Jun Media Hitachi AI Blog
Research
AI Fundamentals
- AI4S · AI4R · AI4Networks · AI4Optics
- Machine Learning · Computer Vision
- Multi-Modal Analysis (Vision, Audio, Language, Sensor)
- Generative AI (Time-series, Image)
- Numerical Simulation / Digital Twin
- 3D Reconstruction / Generation
- LLM / Multi-Modal LLM
AI Applications
- Deep Research (Agentic Research)
- Industrial & Logistics
- Crowd Simulation
- Traffic Analysis
- Elder Care Solutions
- AI for Optics
- Industry LLMs
Journey
Education
2010–2014
Ph.D in Engineering
Hokkaido University, Japan
CSC Scholarship
2008–2010
Master in Engineering
Tianjin University, China
2004–2008
Bachelor in Engineering & Management
Tianjin University, China
2013
Intern
Iowa University, USA
2013
Intern
Seoul National University, Korea
Experience
2025–Present
Researcher
China Telecom Research Institute, China
2014–2025
Chief Researcher
Hitachi China Research Laboratory, China
- CV / AI / AIGC industry applications
- GenAI, LLM, MLM exploration
2014
Research Training
Hitachi Central Research Laboratory, Japan
Projects
Industry Applications
- Public Security Video Surveillance (2016)
- Airport Ground Operation Analysis (2020)
- Multinational Fast Food Visual Ordering (2021)
- Pharma Behavior Monitoring (2022)
- Beverage Safety Management (2023)
- Auto Parts Quality Inspection (2024)
- Logistics Warehouse Behavior Analysis (2024)
- Power Company Safety Compliance (2025)
- Industry Applications of Large Language Models (DeepSeek)
Research Exploration
- Vision-based Public Transport (Chongqing University)
- Public Area Crowd Analysis (Tsinghua)
- Multi-modal Video Understanding (USTB)
- Image-based 3D Reconstruction (USTB)
- AI-based Optical Systems (USTB)
Open Source
fiber-nlse-pinn AI4Optics
PINN solver for the nonlinear Schrödinger equation in optical fiber; SSFM provides the reference field.
warehouse-moc-sensor dataset
Warehouse MoC+Sensor dataset: paired full-body MoCap and real wrist accelerometer (ESWA 2026).
Structured-DGM demo
Inference demo for structured guided diffusion models for industrial defect image generation (KBS 2025).
Selected Papers
Full list →Journals
Sensor-to-Sensor procedural co-learning for sensor-limited human action recognition.
Expert Systems with Applications, 2026
Structured guided diffusion models for industrial defect image generation.
Knowledge-Based Systems, 2025
Query-based neural network for long-range prediction of optical spatio-temporal dynamics in multimode fibers.
Expert Systems with Applications, 2025
Bridge the gap between simulated and real-world data in optical fiber mode decomposition for accuracy improvement: A deep learning-based co-learning framework with visual similarity-based matching.
Expert Systems with Applications, 2024
Global-Shared Text Representation Based Multi-Stage Fusion Transformer Network for Multi-Modal Dense Video Captioning.
IEEE Transactions on Multimedia, 2023
Multisize Patched Spatial-Temporal Transformer Network for Short-and Long-Term Crowd Flow Prediction.
IEEE Transactions on Intelligent Transportation Systems, 2022
Passenger Flow Estimation Based on Convolutional Neural Network in Public Transportation System.
Knowledge-Based Systems, 2017
Conferences
Unconstrained Vision-based Measurement with Open-World Segmentation.
ICCMS 2025