Metraipolis › archive › dc_80775331b01a

Multimodal Hierarchical Mixture-of-Experts for 3D Medical Vision-Language Modeling — arXiv:2611.08185

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.08185
License (as archived): CC0-1.0
Archived: 2026-11-06T14:53:43.943Z
SHA-256 of the archived text: 00e71481f8f5a8cf0314b8787ab06917de4cd5bdaf54630260be9e37568ea48e
Stable URL of this page: https://metraipolis.com/documento/dc_80775331b01a

How to cite

Multimodal Hierarchical Mixture-of-Experts for 3D Medical Vision-Language Modeling — arXiv:2611.08185 — arXiv (en), https://arxiv.org/abs/2611.08185 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_80775331b01a (Metraipolis archive dc_80775331b01a, sha256 00e71481f8f5a8cf0314b8787ab06917de4cd5bdaf54630260be9e37568ea48e)

BibTeX

@misc{dc80775331b01a,
  title = {Multimodal Hierarchical Mixture-of-Experts for 3D Medical Vision-Language Modeling — arXiv:2611.08185},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.08185},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_80775331b01a; sha256 00e71481f8f5a8cf0314b8787ab06917de4cd5bdaf54630260be9e37568ea48e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.08185. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Medical 3D scans and reports are inherently hierarchical, with diagnostic cues concentrated in specific anatomical regions across multiple levels of organization. Yet, most medical vision-language models ignore this structure and fail to focus on clinically relevant anatomy. We introduce MedHiVE, which combines hierarchical vision-language Soft Mixture-of-Experts (SMoE) with contrastive learning to jointly model fine-grained anatomy and global scan context. To our knowledge, MedHiVE is the first framework to route both vision and text tokens through a two-level SMoE hierarchy organized by anatomical scale, coupling this with report-derived anatomy-level contrastive alignment for 3D medical imaging. This replaces the direct region-text alignment of statically pooled features and vision-only hierarchical routing of prior work with dynamic, slot-based cross-modal information flow across anatomical scales. MedHiVE operates via a two-level, learnable token-routing mechanism, where local image and text features are first routed through a local SMoE and then fused by a global SMoE into unified multimodal representations. Contrastive learning aligns both anatomy-level and global vision-language representations in a unified space. Pre-trained on 25,000+ CT scans with anatomical supervision, MedHiVE consistently surpasses 11 state-of-the-art methods across 6 datasets on 3 tasks: segmentation, classification, and detection, demonstrating the importance of hierarchical modeling. The code for MedHiVE is at this http URL .

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2611.08185 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-06T14:25:42.627Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).