Metraipolis › archive › dc_efa5dd79d9ff
Masked Privileged-Information Distillation for Multimodal Skin Lesion Classification Under Missing Clinical Metadata — arXiv:2610.03991
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.03991
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T05:31:56.555Z
- SHA-256 of the archived text
76af2670bbfd9a792dfb4958e805415b9d27e8be7149908032b0e08e25a03444- Stable URL of this page
https://metraipolis.com/documento/dc_efa5dd79d9ff
How to cite
Masked Privileged-Information Distillation for Multimodal Skin Lesion Classification Under Missing Clinical Metadata — arXiv:2610.03991 — arXiv, https://arxiv.org/abs/2610.03991 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_efa5dd79d9ff (Metraipolis archive dc_efa5dd79d9ff, sha256 76af2670bbfd9a792dfb4958e805415b9d27e8be7149908032b0e08e25a03444)
BibTeX
@misc{dcefa5dd79d9ff,
title = {Masked Privileged-Information Distillation for Multimodal Skin Lesion Classification Under Missing Clinical Metadata — arXiv:2610.03991},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.03991},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_efa5dd79d9ff; sha256 76af2670bbfd9a792dfb4958e805415b9d27e8be7149908032b0e08e25a03444},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.03991. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Multimodal skin lesion classification combines clinical images with patient metadata to improve diagnostic accuracy. However, complete metadata available during training may be only partially accessible at deployment, and resource-constrained settings additionally require computational efficiency. We address these challenges with a privileged-information distillation framework in which a multimodal teacher trained on complete metadata supervises a 9.2x smaller student trained with randomly masked clinical fields. Clinical fields are masked as whole groups at a per-sample rate drawn from U(0,1), so one training run covers the full metadata availability range. Fusion is residual, with metadata added as a gated correction to an unconditional image base. On the PAD-UFES-20 dataset, distillation under masked training improves balanced accuracy over cross-entropy training at every availability level, by an average of +4.7 points versus +1.8 points without masking. The masked student loses only 7.8 balanced-accuracy points as metadata decreases from complete to absent, compared with 36.6 points for the same student trained on complete metadata, highlighting the role of masked training in graceful degradation beyond distillation alone. Grad-CAM visualizations further show that the masked student's attention generally remains lesion-centered as metadata is withdrawn. The resulting compact model targets point-of-care settings, where clinical metadata is often incomplete. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2610.03991 — 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 2026-10-06T05:27:05.934Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).