Overview

  • Data Orchestration for AI develops specialized data agents that orchestrate profiling, discovery, assemblage, integration, knowledge building, and visualization over cross-modal, multi-modal data assets (including tables, text, charts, and time series). It optimizes cost, end-to-end time, computing resources, and solution quality. The resulting workflows support downstream model training, question answering, insight generation, and analytical reasoning.
  • Scholarly Data Management develops systems for organizing, analyzing, and utilizing scholarly data, enabling users to store and analyze such data more efficiently. Representative tasks include building taxonomies for incremental and comprehensive scholarly repositories, identifying relevant reviewers for submitted papers, and organizing accepted papers into coherent sessions.
  • Autonomous AI Database System (AI4DB) advances autonomous, learning-enhanced database components—covering learned indexes, cardinality estimation, index advising, and query optimization—designed to work under realistic storage and workload constraints.
  • Civil Computing focuses on interactive visualized data exploration that brings big data back to a human scale for decision-making, spanning domains such as site selection, house seeking, and intelligent transport.
Last updated on May 28, 2026