Exploring Natural Language Interaction Methods for Database Systems

Fan Yuankai – Postdoctoral researcher, Fudan University

Personal introduction: Fan Yuankai is a postdoctoral researcher at Fudan University, holding a Ph.D. in Software Engineering from the same institution. His research focuses on AI for databases (including NL2SQL and query explainability), traditional NLP tasks such as machine translation and semantic parsing, and large‑model techniques like prompt engineering, parameter‑efficient fine‑tuning, tool learning, and model quantization/pruning. He has published seven papers in venues such as ICDE, KBS, WISE, and WWWJ.

Talk Title: Exploring Natural Language Interaction Methods for Database Systems

Talk outline: Providing natural‑language query capabilities for databases can significantly enhance their usability. Users can submit queries in everyday language without needing to know SQL or the underlying data storage structure, obtaining relevant data directly. Recent rapid advances in machine learning, especially in neural machine translation, have produced many translation methods. Building on these advances, the speaker investigates natural‑language interaction techniques from a database perspective.

Audience Benefits:

1. Beyond traditional sequence‑to‑sequence translation methods for NL2SQL, what other approaches exist?

2. How can more reliable ranking mechanisms be used to obtain more trustworthy translation results?

3. How can large models be guided to generate more reliable SQL statements?

Scan the QR code to purchase tickets.