Our paper on fine-grained structured table retrieval from data lakes has been accepted to the aiDM workshop at SIGMOD’24.

The paper takes on two important limitations of existing table retrieval approaches: the returned tables often contain irrelevant data, and the information of interest may be scattered across multiple tables.

To address these issues, we propose the idea of fine-grained structured table retrieval. Whereas existing table retrievers rank the data lake tables according to their relevance to a user-specified query, our goal is to construct small relational databases tailored to user queries.

You may find the paper here.