Experience

Ph.D. Researcher

DAMI-Lab, Rensselaer Polytechnic Institute

Lead first-author projects on multimodal RAG and KB-VQA, studying how VLM/MLLM systems ground visual queries to external knowledge. My recent work examines KB-VQA benchmark reliability, training-free entity identification, and foundation-model evaluation for non-textual structure.
Multimodal RAG KB-VQA VLM Evaluation Graph Foundation Models

Graduate Research Intern

AT&T Chief Data Office

Conducted research on multimodal RAG for knowledge-grounded visual question answering, built reproducible datasets, prompts, evaluation code, and retrieval baselines, and initiated directions that led to first-authored ECCV 2026 and ACL 2026 papers.
Multimodal RAG Knowledge-Grounded VQA Evaluation

Research Assistant

AML Lab, City University of Hong Kong

First-authored HIEST (CIKM 2023), showing that traffic forecasting benefits from modeling regional and global sensor hierarchies. Collaborated on PromptST for prompt-enhanced spatio-temporal multi-attribute prediction.
Traffic Forecasting Time-Series Modeling Urban Computing

Software Engineering Intern, SAP Data Intelligence

SAP Labs China

Developed and maintained two CI/CD automation microservices for a cloud infrastructure service automation platform and worked with SaaS and cloud-native teams on infrastructure workflows.
CI/CD Cloud Infrastructure Microservices

Undergraduate Research Assistant

University of Electronic Science and Technology of China

Processed large-scale urban driving records for grid-based urban computing research and implemented MapReduce-style parallel preprocessing.
Urban Computing Data Preprocessing

Education

Ph.D. Student in Computer Science

Rensselaer Polytechnic Institute

Advised by Prof. Yao Ma.

M.Sc. in Multimedia Information Technology, Distinction

City University of Hong Kong

B.Eng. in Software Engineering - Systems and Technology

University of Electronic Science and Technology of China

Technical Skills
Programming
Python
PyTorch / PyTorch Lightning
Java / JavaScript / Go / C
Multimodal and RAG
VLM/MLLM evaluation
Qwen-VL, LLaVA, Llama, EVA-CLIP, CLIP
FAISS, BGE reranking, prompt-based evaluation
Graphs and Time Series
GNNs and graph self-supervised learning
Graph contrastive learning and graph foundation models
Spatio-temporal forecasting
Systems and Tools
Linux, Git, Docker, Kubernetes
Slurm, Singularity, Jenkins
LaTeX
Honors and Awards
Outstanding Student Scholarship
University of Electronic Science and Technology of China ∙ 2021
Outstanding Student Scholarship
University of Electronic Science and Technology of China ∙ 2020
Outstanding Student Scholarship
University of Electronic Science and Technology of China ∙ 2018
Languages
100%
Mandarin Native
90%
English Fluent; IELTS 7.0