[ivory-search id="3463" title="Default Search Form"]
Search
Buy your pass
Search
Type below to search:
OnlineDiscovery - From Molecules to Models Badge available

What AI is leaving on the table: Towards democratizing insights from tabular data

  • Register

    * Register (or log in) to the Neural Network to add this session to your agenda or watch the replay

    • Date
    23 February 2026
    Timeframe
    16:00 - 17:00 CET Geneva
    Duration
    60 minutes
    • Days
      Hours
      Min
      Sec

    With the rise of neural models pretrained on massive datasets, text and images have become first-class citizens in AI. Tabular data, common in relational databases and spreadsheets, has long been overlooked, despite remaining the dominant data modality in most organizations. Recently, “Tabular AI” has emerged as a key area of interest and has proven important for democratizing valuable insights from tabular data.

    In the first part of the seminar, the speaker introduces neural models for tabular data and reviews key properties of tabular data that such models should capture. The seminar then covers different approaches to inferring table semantics, a cornerstone of many tabular analysis tasks. It further presents a framework that leverages table semantics to detect context-sensitive data in tables, enabling safer sharing of tabular datasets.

    In the second part, the seminar takes an end-to-end view of data analysis pipelines, beginning with two perspectives on table retrieval. It discusses AI-powered methods for explicit dataset search as well as end-to-end insight extraction pipelines, where table retrieval is more implicit. It then turns to AI-powered tabular reasoning, from LLM-based approaches to text-to-SQL, for answering analytical questions in natural language.

    In the final part, the seminar reflects on the future of tabular data science systems and processes, and on the emerging capabilities data scientists will need to unlock the full value of data science.

     

    Session Objectives:

    By the end of this session, participants will be able to:

    • Understanding transformer-models for tabular data and how they can be applied to surface table semantics.
    • Applying tabular AI to table retrieval, processing, and querying in end-to-end pipelines for insight extraction from tabular data.
    • Reflecting on a possible future for tabular data science and the critical skills that are required of the next-generation data analysts and data scientists.

     

    Recommended Mastery Level / Prerequisites:

    • A basic understanding of machine learning and deep learning.
    • Familiarity with data science processes.
    Share this session with your network
    • Session starts in
      Days
      Hours
      Min
      Sec
    Add to Calendar
    Discover more from programme stream
    Discovery - From Molecules to Models
    The why of AI: Uncovering cause and effect in observational...

    The why of AI: Uncovering cause and effect in observational...

    18 February 2026 - 16:00 to 17:00

    What AI is leaving on the table: Towards democratizing insights...

    What AI is leaving on the table: Towards democratizing insights...

    23 February 2026 - 16:00 to 17:00

    Related stories

    Are you sure you want to remove this speaker?