4 papers
SEGA: Spectral-Energy Guided Attention for Resolution Extrapolation in Diffusion Transformers
Javad Rajabi, Kimia Shaban, Koorosh Roohi +2
Diffusion transformers (DiTs) have emerged as a dominant architecture for text-to-image generation, yet their performance drops when generating at resolutions beyond their training…
Sizes of witnesses in Covtree
Jette Gutzeit, Kimia Shaban, Karen Yeats +1
Given a set of unlabelled posets, each of size , we say that a poset is a \emph{witness} to if is the set of downsets of size of . We say that …
When Does RL Help Medical VLMs? Disentangling Vision, SFT, and RL Gains
Ahmadreza Jeddi, Kimia Shaban, Negin Baghbanzadeh +4
Reinforcement learning (RL) is increasingly used to post-train medical Vision-Language Models (VLMs), yet it remains unclear whether RL improves medical visual reasoning or mainly…
Statistics and asymptotics of subdivergence-free Feynman integrals in theory
Paul-Hermann Balduf, Kimia Shaban, Johannes Thürigen
Recent algorithmic improvements have made it possible to evaluate subdivergence-free (=primitive=skeleton) Feynman integrals in theory numerically up to 18 loops. By now, al…