Metraipolis › archive › dc_8ec57044ca80
LaSeD: Label-Semantic Self-Distillation for Visual-Only Surgical Phase Recognition — arXiv:2609.18971
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.18971
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T16:16:46.189Z
- SHA-256 of the archived text
a80ee5a771b144a4e95d92fa6666ce9e6f3b88609352b81e1bf1d7a4c2a1c639- Stable URL of this page
https://metraipolis.com/documento/dc_8ec57044ca80
How to cite
LaSeD: Label-Semantic Self-Distillation for Visual-Only Surgical Phase Recognition — arXiv:2609.18971 — arXiv (en), https://arxiv.org/abs/2609.18971 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8ec57044ca80 (Metraipolis archive dc_8ec57044ca80, sha256 a80ee5a771b144a4e95d92fa6666ce9e6f3b88609352b81e1bf1d7a4c2a1c639)
BibTeX
@misc{dc8ec57044ca80,
title = {LaSeD: Label-Semantic Self-Distillation for Visual-Only Surgical Phase Recognition — arXiv:2609.18971},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.18971},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8ec57044ca80; sha256 a80ee5a771b144a4e95d92fa6666ce9e6f3b88609352b81e1bf1d7a4c2a1c639},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.18971. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Surgical phase recognition maps each video frame to a clinically meaningful workflow phase, supporting context-aware assistance, documentation, and postoperative analysis. Most methods treat phase annotations only as class IDs, whereas recent surgical vision-language models often require additional video--text data, captions, or instruction tuning. We propose \emph{LaSeD}, a label-semantic self-distillation framework that uses phase names as privileged training-time context while retaining visual-only deployment without a ground-truth phase-name hint. LaSeD initializes a frozen teacher and a student from the same pretrained VLM checkpoint. The teacher receives the frame, a fixed task prompt, and the ground-truth phase-name hint; the student receives the same frame and prompt without the hint, and only its visual encoder is optimized. Training combines hard phase-token supervision with feature-level distillation from cached teacher representations. At inference, the teacher and hint are removed, and the student predicts one of the seven Cholec80 phases through constrained digit-token logits without an additional classifier head. On the Cholec80 evaluation split, LaSeD achieves 86.20\% accuracy, 77.75\% macro recall, 78.13\% macro precision, and 64.15\% macro Jaccard. Under the identical protocol, it improves a visual-only Qwen3-VL-4B baseline by 9.45, 9.42, 11.93, and 12.22 percentage points, respectively. Visual-only means that the image is the only sample-specific inference input, while all frames share the same fixed task prompt. These results suggest that phase names provide a useful low-cost signal for adapting VLMs to surgical workflow analysis. (The code will be published soon.)
Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-17 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.18971 — 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 2026-10-15T15:45:19.831Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).