Paolo Papotti Keynote Speaker at LLIS workshop on Representation Learning and Generative Models for Structured Data

Machine Learning is well-established in areas like text, images, audio, and video, but structured data (e.g., relational tables and knowledge graphs) has received less attention. Recent studies explore integrating this data with ML models. This workshop focused on representation learning and generative models for structured data, including knowledge graphs, and also covered the intersection of structured data learning and information retrieval, such as retrieval-augmented generation (RAG) and question answering (QA). The goal was to connect researchers from NLP, ML, DB, and IR to foster new ideas and collaboration.

On this occasion, Paolo Papotti, 3IA Côte d'Azur Chairholder, gave a conference entitled "SQL and Large Language Models: A Marriage Made in Heaven?".
Asbtract: With the rise of pre-trained Large Language Models (LLMs), there is now an effective solution to store and use information extracted from massive corpora of documents. However, for data-intensive tasks over structured data, relational DBs and SQL queries are at the core of countless applications. While these two technologies may appear distant, in this talk we will see that they can interact effectively and with promising results. LLMs can help users express SQL queries (Semantic Parsing), but SQL queries can be used to evaluate LLMs (Benchmarking). Their combination can be further advanced, with opportunities to query with a unified SQL interface both LLMs and DBs. We present recent results on these topics and then conclude with an overview of the research challenges in effectively leveraging the combined power of SQL and LLMs.

Learn more about this event