AI4DB

We apply AI/ML techniques to optimize and enhance database systems.

Subtopics

1. Query Optimization

• RankPQO: Learning-to-Rank for Parametric Query Optimization.
  Songsong Mo, Yue Zhao, Zhifeng Bao, Quanqing Xu, Chuanhui Yang, Gao Cong.
  PVLDB 2025.

• LLM-R2: A Large Language Model Enhanced Rule-based Rewrite System for Boosting Query Efficiency.
  Zhaodonghui Li, Haitao Yuan, Huiming Wang, Gao Cong, Lidong Bing.
  PVLDB 2025.

2. ML-enhanced indexes

• BT-Tree: A Reinforcement Learning Based Index for Big Trajectory Data.
  Tu Gu, Kaiyu Feng, Jingyi Yang, Gao Cong, Long Cheng, Rui Zhang.
  SIGMOD 2025.

• MAAdvisor: Zero-Shot Index Advisor using Multi-Agent LLMs.
  Zhaodonghui Li, Haitao Yuan, Jiachen Shi, Hao Zhang, Yu Rong, Gao Cong.
  CoRR abs/2508.16044 (2025).

3. Query Representation Learning

• TATA: An Efficient Framework for Task Transfer in Query Plan Representation.
  Yue Zhao, Songsong Mo, Gao Cong.
  PVLDB 2026.