Picture for Thomas Hofmann

Thomas Hofmann

ETH Zurich

Scalable Non-Equivariant 3D Molecule Generation via Rotational Alignment

Add code
Jun 11, 2025
Viaarxiv icon

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

Add code
May 25, 2025
Viaarxiv icon

Generalized Interpolating Discrete Diffusion

Add code
Mar 06, 2025
Viaarxiv icon

IC-Portrait: In-Context Matching for View-Consistent Personalized Portrait

Add code
Jan 31, 2025
Viaarxiv icon

SHYI: Action Support for Contrastive Learning in High-Fidelity Text-to-Image Generation

Add code
Jan 15, 2025
Figure 1 for SHYI: Action Support for Contrastive Learning in High-Fidelity Text-to-Image Generation
Figure 2 for SHYI: Action Support for Contrastive Learning in High-Fidelity Text-to-Image Generation
Figure 3 for SHYI: Action Support for Contrastive Learning in High-Fidelity Text-to-Image Generation
Figure 4 for SHYI: Action Support for Contrastive Learning in High-Fidelity Text-to-Image Generation
Viaarxiv icon

On the Expressiveness and Length Generalization of Selective State-Space Models on Regular Languages

Add code
Dec 26, 2024
Figure 1 for On the Expressiveness and Length Generalization of Selective State-Space Models on Regular Languages
Figure 2 for On the Expressiveness and Length Generalization of Selective State-Space Models on Regular Languages
Figure 3 for On the Expressiveness and Length Generalization of Selective State-Space Models on Regular Languages
Figure 4 for On the Expressiveness and Length Generalization of Selective State-Space Models on Regular Languages
Viaarxiv icon

UIP2P: Unsupervised Instruction-based Image Editing via Cycle Edit Consistency

Add code
Dec 19, 2024
Figure 1 for UIP2P: Unsupervised Instruction-based Image Editing via Cycle Edit Consistency
Figure 2 for UIP2P: Unsupervised Instruction-based Image Editing via Cycle Edit Consistency
Figure 3 for UIP2P: Unsupervised Instruction-based Image Editing via Cycle Edit Consistency
Figure 4 for UIP2P: Unsupervised Instruction-based Image Editing via Cycle Edit Consistency
Viaarxiv icon

LoRACLR: Contrastive Adaptation for Customization of Diffusion Models

Add code
Dec 12, 2024
Viaarxiv icon

MegaPortrait: Revisiting Diffusion Control for High-fidelity Portrait Generation

Add code
Nov 07, 2024
Viaarxiv icon

Emergence of Globally Attracting Fixed Points in Deep Neural Networks With Nonlinear Activations

Add code
Oct 29, 2024
Viaarxiv icon