Metraipolis › archive › dc_dee0095185d6

What to Preserve, Where to Adapt: A Depth-Wise Analysis of Forgetting in Continual Gynecological Image Segmentation — arXiv:2608.13660

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.13660
License (as archived)
CC0-1.0
Archived
2026-10-19T02:30:44.862Z
SHA-256 of the archived text
4ade5d61f26c875423429c750ee7adde4371469086fe6ce14dafa170a856d60e
Stable URL of this page
https://metraipolis.com/documento/dc_dee0095185d6

How to cite

What to Preserve, Where to Adapt: A Depth-Wise Analysis of Forgetting in Continual Gynecological Image Segmentation — arXiv:2608.13660 — arXiv (en), https://arxiv.org/abs/2608.13660 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_dee0095185d6 (Metraipolis archive dc_dee0095185d6, sha256 4ade5d61f26c875423429c750ee7adde4371469086fe6ce14dafa170a856d60e)
BibTeX
@misc{dcdee0095185d6,
  title = {What to Preserve, Where to Adapt: A Depth-Wise Analysis of Forgetting in Continual Gynecological Image Segmentation — arXiv:2608.13660},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.13660},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_dee0095185d6; sha256 4ade5d61f26c875423429c750ee7adde4371469086fe6ce14dafa170a856d60e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.13660. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

The clinical management of gynecological diseases often relies on medical imaging for diagnosis, treatment planning, and follow-up. Segmentation in this setting is challenging because successive tasks may differ in imaging modality, target anatomy, pathology, and annotation structure. Continual learning allows models to adapt to new tasks without simultaneous access to previous datasets. However, when successive tasks differ substantially, learning a new task can degrade performance on earlier ones, a problem known as catastrophic forgetting. Understanding where adaptation disrupts previous knowledge can help guide the design of more targeted continual-learning strategies. We investigate how forgetting changes as different parts of an encoder--decoder network are allowed to adapt. We progressively expand the trainable region of a 3D nnU-Net backbone from the bottleneck toward input- and output-proximal blocks. Under a shared learning rate, adaptation near the bottleneck largely preserves previous-task performance but provides limited current-task learning, whereas broader adaptation improves current-task performance but sharply increases forgetting. This trade-off persists even when the average change in trainable backbone parameters is approximately comparable. Assigning different learning rates to different blocks substantially reduces forgetting when part of the backbone is trainable, although this changes both the size and location of the updates. Forgetting still increases as more blocks are trained and remains severe when the full backbone is updated. These results show that forgetting depends not only on how much the model changes, but also on which parts of the model are allowed to change.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-17 in cs.LG).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.13660 — 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 2026-10-19T02:26:17.818Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).