Metraipolis › archive › dc_ba81769cd7cb
Parameter-Efficient Fine-Tuning of Foundation Models for Liver Tumor Segmentation in CT — arXiv:2609.14106
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.14106
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T01:52:53.742Z
- SHA-256 of the archived text
3d05d70f59b6c460111fc7849124b1d875312cf8b033c1fb3315338bad915314- Stable URL of this page
https://metraipolis.com/documento/dc_ba81769cd7cb
How to cite
Parameter-Efficient Fine-Tuning of Foundation Models for Liver Tumor Segmentation in CT — arXiv:2609.14106 — arXiv (en), https://arxiv.org/abs/2609.14106 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ba81769cd7cb (Metraipolis archive dc_ba81769cd7cb, sha256 3d05d70f59b6c460111fc7849124b1d875312cf8b033c1fb3315338bad915314)
BibTeX
@misc{dcba81769cd7cb,
title = {Parameter-Efficient Fine-Tuning of Foundation Models for Liver Tumor Segmentation in CT — arXiv:2609.14106},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.14106},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ba81769cd7cb; sha256 3d05d70f59b6c460111fc7849124b1d875312cf8b033c1fb3315338bad915314},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.14106. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We evaluated parameter-efficient fine-tuning (PEFT) of the Segment Anything Model (SAM) for liver tumor segmentation in abdominal CT of colorectal liver metastases. We compared Low-Rank Adaptation (LoRA), 4-bit Quantized LoRA (QLoRA), a convolutional adapter (Conv-Adapter), and our Directional Spectral Top-K adapter (DiSCo), training only adapters while freezing the SAM backbone. DiSCo derives spectral bases from singular value decomposition of row-normalized weights and learns rank-gated spectral coefficients, per-output magnitude offsets, and a spectral gain, with optional Top-K rank selection at inference and 0.14 M trainable parameters. We benchmarked five prompting regimes: no prompt, single-point, multi-point, and bounding boxes at intersection over union 0.50 and 0.75. Conv-Adapter and LoRA achieved the highest accuracy (overall Dice 0.793 and 0.792; single-point Dice 0.795 and 0.792; 95th-percentile Hausdorff distance (HD95) 32 mm). QLoRA was close (overall Dice 0.766; single-point Dice 0.768; HD95 36.41 mm), with 0.91 M trainable parameters, 120 ms latency, and 4.9 GB peak memory. DiSCo achieved the highest Dice per million trainable parameters (4.66), with overall Dice 0.653, single-point Dice 0.698, and HD95 49.53 mm. These results show an accuracy-efficiency trade-off and support PEFT for liver tumor segmentation with reduced adaptation costs when compute and labeled data are limited. Code: this https URL Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-15 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.14106 — 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 2026-10-16T01:25:23.557Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).