Eric Tillmann Bill • Home

Hi, I’m Eric! 👋

I’m a master’s student at ETH Zurich working on generative image models. I began at RWTH Aachen, where I wrote my bachelor’s thesis on graph neural networks with Christopher Morris. From there, I shifted into generative modeling at ETH Zurich, exploring magnitude-preservation methods for Diffusion Transformers with Cristian Perez Jensen (NeurIPS’26 OPT). Recently, I’ve focused on multi-subject text-to-image: JEDI (ICML’25 PUT), followed by FOCUS, which casts multi-subject fidelity as stochastic optimal control and improves on JEDI, with Enis Simsar and Thomas Hofmann.

I’ve completed industry internships in Stuttgart and Singapore, and I’m now finishing my master’s thesis on vision-language models via discrete flow matching. I expect to graduate in April 2026.

News

Sep 2025, Map-DiT accepted to NeurIPS'25 Workshop on Optimization for ML 🧶
Jul 2025, JEDI accepted to ICML’25 Workshop on Test-Time Adaptation 🔬

Selected Publications

Optimal Control Meets Flow Matching: A Principled Route to Multi-Subject Fidelity

Eric Tillmann Bill, Enis Simsar, Thomas Hofmann

Preprint ArXiv

Preprint Code Webpage

BibTeX

@misc{bill2025focus,
       title={Optimal Control Meets Flow Matching: A Principled Route to Multi-Subject Fidelity}, 
       author={Eric Tillmann Bill and Enis Simsar and Thomas Hofmann},
       year={2025},
       eprint={2510.02315},
       archivePrefix={arXiv},
       primaryClass={cs.CV},
       url={https://arxiv.org/abs/2510.02315}, 
 }

Exploring Magnitude Preservation and Rotation Modulation in Diffusion Transformers

Eric Tillmann Bill, Cristian Perez Jensen, Sotiris Anagnostidis, Dimitri von Rütte

NeurIPS 2025, Optimization for Machine Learning Workshop

Preprint OpenReview Code

BibTeX

@inproceedings{bill2025exploring,
	title={Exploring Magnitude Preservation and Rotation Modulation in Diffusion Transformers},
	author={Anonymous},
	booktitle={OPT 2025: Optimization for Machine Learning},
	year={2025},
	url={https://openreview.net/forum?id=j5NB2cpxIQ}
	}

JEDI: The Force of Jensen-Shannon Divergence in Disentangling Diffusion Models

Eric Tillmann Bill, Enis Simsar, Thomas Hofmann

ICML 2025, Putting Updates to the Test Workshop

Preprint OpenReview Code Webpage

BibTeX

@inproceedings{
	 bill2025jedi,
	 title={{JEDI}: The Force of Jensen-Shannon Divergence in Disentangling Diffusion Models},
	 author={Eric Tillmann Bill and Enis Simsar and Thomas Hofmann},
	 booktitle={Second Workshop on Test-Time Adaptation: Putting Updates to the Test! at ICML 2025},
	 year={2025},
	 url={https://openreview.net/forum?id=HVQ3wL2jPI}
}