Learning from Noisy Prompts: Saliency-Guided Prompt Distillation for Robust Segmentation with SAM

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kang, Jingxuan, Zhang, Ziqi, Zheng, Shaoming, Li, Shuang, Patel, Uday Bharat, Fitzhugh, Alexander Harry, Lung, Phillip, Kiberu, Yusuf, Jathanna, Nikesh, Jamil-Copley, Shahnaz, Kainz, Bernhard, Qin, Chen
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910166060892160
author Kang, Jingxuan
Zhang, Ziqi
Zheng, Shaoming
Li, Shuang
Patel, Uday Bharat
Fitzhugh, Alexander Harry
Lung, Phillip
Kiberu, Yusuf
Jathanna, Nikesh
Jamil-Copley, Shahnaz
Kainz, Bernhard
Qin, Chen
author_facet Kang, Jingxuan
Zhang, Ziqi
Zheng, Shaoming
Li, Shuang
Patel, Uday Bharat
Fitzhugh, Alexander Harry
Lung, Phillip
Kiberu, Yusuf
Jathanna, Nikesh
Jamil-Copley, Shahnaz
Kainz, Bernhard
Qin, Chen
contents Segmentation is central to clinical diagnosis and monitoring, yet the reliability of modern foundation models in medical imaging still depends on the availability of precise prompts. The Segment Anything Model (SAM) offers powerful zero-shot capabilities, although it collapses under the weak, generic, and noisy prompts that dominate real clinical workflows. In practice, annotations such as centerline points are coarse and ambiguous, often drifting across neighboring anatomy and misguiding SAM toward inconsistent or incomplete masks. We introduce SPD, a Saliency-Guided Prompt Distillation framework that converts these unreliable cues into robust guidance. SPD first learns data-driven anatomical priors through a lightweight saliency head to obtain confident localization maps. These priors then drive Contextual Prompt Distillation, which validates and enriches noisy prompts using cues from anatomically adjacent slices, producing a consensus prompt set that matches the behavior of expert reasoning. A Pairwise Slice Consistency objective further enforces local anatomical coherence during segmentation. Experiments on four challenging MRI and CT benchmarks demonstrate that SPD consistently outperforms existing SAM adaptations and supervised baselines, delivering large gains in both region-based and boundary-based metrics. SPD provides a practical and principled path toward reliable foundation model deployment in clinical environments where only imperfect prompts are available.
format Preprint
id arxiv_https___arxiv_org_abs_2604_23314
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Learning from Noisy Prompts: Saliency-Guided Prompt Distillation for Robust Segmentation with SAM
Kang, Jingxuan
Zhang, Ziqi
Zheng, Shaoming
Li, Shuang
Patel, Uday Bharat
Fitzhugh, Alexander Harry
Lung, Phillip
Kiberu, Yusuf
Jathanna, Nikesh
Jamil-Copley, Shahnaz
Kainz, Bernhard
Qin, Chen
Computer Vision and Pattern Recognition
Segmentation is central to clinical diagnosis and monitoring, yet the reliability of modern foundation models in medical imaging still depends on the availability of precise prompts. The Segment Anything Model (SAM) offers powerful zero-shot capabilities, although it collapses under the weak, generic, and noisy prompts that dominate real clinical workflows. In practice, annotations such as centerline points are coarse and ambiguous, often drifting across neighboring anatomy and misguiding SAM toward inconsistent or incomplete masks. We introduce SPD, a Saliency-Guided Prompt Distillation framework that converts these unreliable cues into robust guidance. SPD first learns data-driven anatomical priors through a lightweight saliency head to obtain confident localization maps. These priors then drive Contextual Prompt Distillation, which validates and enriches noisy prompts using cues from anatomically adjacent slices, producing a consensus prompt set that matches the behavior of expert reasoning. A Pairwise Slice Consistency objective further enforces local anatomical coherence during segmentation. Experiments on four challenging MRI and CT benchmarks demonstrate that SPD consistently outperforms existing SAM adaptations and supervised baselines, delivering large gains in both region-based and boundary-based metrics. SPD provides a practical and principled path toward reliable foundation model deployment in clinical environments where only imperfect prompts are available.
title Learning from Noisy Prompts: Saliency-Guided Prompt Distillation for Robust Segmentation with SAM
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2604.23314