3 papers
cs.CL2025
Do Depth-Grown Models Overcome the Curse of Depth? An In-Depth Analysis
Ferdinand Kapl, Emmanouil Angelis, Tobias Höppe +4
Gradually growing the depth of Transformers during training can not only reduce training cost but also lead to improved reasoning performance, as shown by MIDAS (Saunshi et al., 20…
stat.ML2025
Minimum-Excess-Work Guidance
Christopher Kolloff, Tobias Höppe, Emmanouil Angelis +4
We propose a regularization framework inspired by thermodynamic work for guiding pre-trained probability flow generative models (e.g., continuous normalizing flows or diffusion mod…
cs.LG2025
In-silico biological discovery with large perturbation models
Djordje Miladinovic, Tobias Höppe, Mathieu Chevalley +6
Data generated in perturbation experiments link perturbations to the changes they elicit and therefore contain information relevant to numerous biological discovery tasks -- from u…