The Distinguished Lecture Series, hosted by the School of Computing and Data Science (CDS), brings leading scholars from around the world to share their cutting-edge research and insights in the fields of computer science, data science, artificial intelligence, and statistics.

We are pleased to announce the 5th lecture in the series, titled “An LLM-Powered Data Analytics System”, to be delivered by Professor Guoliang LI from the Department of Computer Science at Tsinghua University, Beijing, China.

Speaker:

Professor Guoliang LI, Department of Computer Science, Tsinghua University

Moderator:

Professor Reynold Cheng

Date:

22 May, 2025 (Thursday)

Time:

10:30 am – 11:30 am

Venue:

CB-A, G/F, Chow Yei Ching Building, Main Campus, The University of Hong Kong

Abstract:
Data analytics systems for structured data are widely used and deployed. However, analyzing unstructured and heterogeneous data—such as data lakes—remains a significant challenge due to the absence of semantic operators, intelligent data analytics pipeline generation, and effective reasoning capabilities.

Fortunately, large language models (LLMs) come equipped with powerful abilities in understanding, reasoning, semantic matching, and generation, offering an opportunity to transform data analytics systems fundamentally. Firstly, in the context of structured data analytics, LLMs can be integrated as semantic operators within data processing workflows. Secondly, for unstructured data, LLMs can automatically generate execution pipelines to facilitate analysis. Thirdly, when dealing with heterogeneous data, we demonstrate methods to link disparate data types and amalgamate their execution plans.

In this talk, he will explore these challenges and propose solutions to address them. Furthermore, he will highlight the ongoing challenges in the domain of heterogeneous data analytics.

Biography:

Guoliang Li is a full professor in the Department of Computer Science at Tsinghua University, Beijing, China. He is an ACM Fellow and an IEEE Fellow. His research interests include Data+AI Systems, AI4Data, Data4AI, and cloud-native database systems. He has received several awards, including the VLDB 2017 Early Research Contribution Award, TCDE 2014 Early Career Award, SIGMOD 2024 Research Highlight Award, as well as best paper awards, such as VLDB 2023 Best Industry Paper, CIKM 2017 Best Paper award, DASFAA 2023 Best Paper award, best papers of SIGMOD 2023, VLDB 2020, KDD 2018, ICDE 2018. Guoliang has served as the general co-chair of SIGMOD 2021, demo co-chair for VLDB 2022, industrial co-chair for ICDE 2022, tutorial co-chair for SIGMOD 2022, and program contest co-chair for SIGMOD 2024. He regularly serves as a (senior) PC member for conferences like SIGMOD, VLDB, and ICDE.

All are welcome to attend.