Menu

Seminar on Foundation Models for Structured Data

Tabular data modeling is undergoing rapid transformation driven by foundation models. This seminar examines the most recent advances, with a focus on state-of-the-art approaches across i.i.d., time-series, and causal reasoning settings.

Kick-Off :

Course type:Seminar
Time- Kick-off Session: 22.04.2026 - 10:30 am (Zoom)
- Presentation Session 1: 12.06.2026, 10-12am
- Presentation Session 2: 19.06.2026, 10-12am
- Presentation Session 3: 26.06.2026, 10-12am
- Presentation Session 4: 03.07.2026, 10-12am
- Presentation Session 5: 10.07.2026, 10-12am
- Presentation Session 6: 17.07.2026, 10-12am
LocationIn-Person; Room SR 00-010/14, Building 101
OrganizersMagnus Bühler, Jake Robertson, Omar Swelam, Liam and Steven Adriaensen and Frank Hutter
RegistrationOpen
LanguageEnglish

Prerequisites

We require that you have taken lectures on or are familiar with the following:

  • Machine Learning
  • Deep Learning

The seminar covers tabular foundation models with a focus on time series and causal inference and discovery. Pointers to external resources will be provided for self-study.

Organization

  • 24 Students (2 students per group & 2 groups per session)
  • Each team gets assigned a buddy team
    • Dry run presentation with feedback
  • Presentations in pairs
    • Paper presentation (methodology, evaluation, results,...)
    • Simulate a conference reviewer (critical analysis of the paper)
    • Additional Evaluations, extensions,...
  • Kahoot Quiz

Grading

  • Presentation

Papers

Tabular Foundation Models:

Time - Series:

Causal Reasoning:

  • 10.07.2026 - 10-11am: [Group 9] Avici
  • 10.07.2026 - 11-12am: [Group 10] BCNP
  • 17.07.2026 - 10-11am: [Group 11] Do-PFN
  • 17.07.2026 - 11-12am: [Group 12] Use What You Know