Multimodal Retrieval-Augmented Generation for Financial Documents
A multimodal RAG approach for financial documents that combines chart and table parsing, Markdown conversion, and hybrid retrieval to improve document understanding and generation.
01
Research Interests
Retrieval-Augmented Generation
Coding Agents
02
Publications
03
Experience
Guangdong Singularity.ai
Developed LLM-powered e-commerce livestreaming systems, including speech data processing, Qwen fine-tuning, and ASR/TTS pipeline optimization.
LinkTime Cloud
Conducted research and engineering on RAG systems, including hybrid retrieval, Graph RAG, and evaluation frameworks.
04
Projects
No projects listed yet.
05
Education
M.S. in Software Engineering
Sun Yat-sen University, School of Software Engineering
B.Eng. in Data Science and Big Data Technology
Wuhan Textile University, School of Computer Science and Artificial Intelligence
06