Reproduce “Multimodal Hierarchical Mixture-of-Experts for 3D Medical Vision-Language Modeling” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Gizem Mert, Szymon Płotka, Pedro R. A. S. Bassi, Wenxuan Li, Zongwei Zhou, Jarosław B. Ćwikła, Arkadiusz Sitek, Ewa Szczurek

MedHiVE combines image and report features through organ-level and whole-scan routing, then fuses them for 3D segmentation. On the authors’ KiTS split of public training data, it reaches 66.7% mDSC, the mean class-wise Dice score, versus 60.1% for MambaHoME. Removing second-level routing lowers four-benchmark aggregate mDSC from 80.2% to 54.5%; these experiments do not establish clinical readiness.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.