Raphaël Baur

Doctoral Fellow, ETH AI Center

ETH Zürich

I am a Doctoral Fellow at the ETH AI Center. My research is supervised by Prof. Dr. Mennatallah El-Assady and Prof. Dr. Christoph Hölscher.

I am interested to use human-AI collaboration to integrate human expertise and creativity into complex design processes, in particular in the field of hospital designs.

To achieve this, I am currently focusing on capturing complex human preferences by learning reward functions from heterogenous human feedback.

News

  • MAVRL has been accepted to ICML 2026. See you all in Seoul!
  • We just published a preprint on MAVRL, a novel method for learning reward functions from multiple feedback types with amortized variational inference.
  • Pre-alpha release of arxitect, a web-based layout-generation tool with real-time, human-centred analytics
  • I have joined the ETH AI Center as a Doctoral Fellow under the supervision of Prof. Menna El-Assady and Prof. Christoph Hölscher
  • I have successfully defended my Master Thesis supervised by Prof. Menna El-Assady and Prof. Christoph Hölscher
  • The EBD-Toolkit has received a major update, including added visualization functionality and dedicated tutorials
  • We kicked off the course Evidence-Based Design in Architecture, where I will be TAing and teaching this semester
  • I started working as a research assistant at the IVIA Lab

Selected Publications

MAVRL: Learning Reward Functions from Multiple Feedback Types with Amortized Variational Inference
Raphaël Baur, Yannick Metz, Maria Gkoulta, Mennatallah El-Assady, Giorgia Ramponi, Thomas Kleine Buening
ICML2026
The DesignMind Toolkit: A Toolkit for Evidence-Based, Cognitively-Informed and Human-Centred Architectural Design
Michal Gath-Morad, Raphaël Baur, Christoph Hölscher
ECAADE2023
RLHF-Blender: A Configurable Interactive Interface for Learning from Diverse Human Feedback
Yannick Metz, David Lindner, Raphaël Baur, Daniel Keim, Mennatallah El-Assady
ICML2023