About

I am a graduate student at the College of Design and Engineering, National University of Singapore, pursuing a Master's degree in Industrial Systems Engineering and Management (Data Analytics track). I received my B.S. in Information Management and Information Systems from the College of Information Science and Technology, Nanjing University of Aeronautics and Astronautics, in 2025.

My research interests include Multimodal Large Language Models, AI agents, recommender systems, and retrieval-augmented generation (RAG). I am dedicated to exploring the broader capabilities of large language models and new methods for multimodal information fusion.

If you are interested in collaboration or mentorship discussions, please feel free to contact me by email.

Research Interests

Multimodal Large Language Models
AI Agent
Recommender Systems
Retrieval-Augmented Generation (RAG)

Education

National University of Singapore

M.Sc in Industrial Systems Engineering and Management (Data Analytics), Aug 2025 - Jun 2026
Core Courses: Industrial System Programming, Business Intelligence, Recommender Systems, Graph Processing, Applied System Modeling

Nanjing University of Aeronautics and Astronautics

B.S. in Information Management and Information Systems, Sep 2021 - Jun 2025
Core Courses: Data Analysis and Applications, Enterprise Digital Principles, Python Data Analysis, Data Structures

Research & Internship Experience

Oct 2025 - Present

Research Assistant @ Tsinghua University, School of Economics and Management

Working under Prof. Ning Li on simulated dialogue practice for real business scenarios. Built a data pipeline from JSON to CSV with comprehensive metrics including turn/role distribution, vocabulary diversity (TTR), sentiment and terminology ratios. Integrated LLM for key point scoring and robust parsing. Implemented multi-threading, caching, error handling and logging for efficient batch processing and report generation.

Supervisor: Prof. Ning Li

Aug 2025 - Present

Research Intern @ National University of Singapore, School of Computing

Participating in building PredictBench, a predictive spatial reasoning benchmark. Leading Task1-VQA design and evaluation pipeline, segmenting videos into "visible input clips + hidden GT clips" to ensure questions require predictive inference with unique answers. Collaborating on segment validation rules, baselines, ablations and evaluation scripts. Project in model testing phase, planned for CVPR submission in November (2nd contributor).

Dec 2024 - May 2025

Information System Project @ NUAA Research Institute

Developed information collection platform for faculty members, implementing unified aggregation of personal information. Responsible for Vue frontend development using Ant Design Pro and platform data maintenance.

Publications

Counterfactual Causal Inference with Multi-Grained Contrastive Representation Learning for Popularity-Debiased Multi-modal Bundle Recommendation

Yang Jie, et al.

Submitted

This paper focuses on multimodal bundle recommendation, addressing popularity bias through counterfactual causal inference and multi-grained contrastive representation learning.

Cross-modal RAG Framework for Image-Text Retrieval Augmented Generation

Yang Jie

Bachelor's Thesis Nanjing University of Aeronautics and Astronautics

Design and implementation of an image-text retrieval augmented generation system based on LangChain framework. Systematic comparison of Google Gemini native generation, simple RAG model and the proposed cross-modal RAG framework shows significant advantages in reasoning accuracy, semantic consistency and response relevance. Proposed cross-modal semantic indexing mechanism based on vector database and hybrid retrieval strategy, experimentally validated the model's advantages in cross-modal information fusion and logical reasoning coherence.