Saved in:
Bibliographic Details
Main Authors: Lu, Zhixiang, Xu, Shijie, Yan, Kaicheng, Cai, Xuyue, Zhang, Chong, Li, Yulong, Stefanidis, Angelos, Nguyen, Anh, Su, Jionglong
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.21010
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910062759378944
author Lu, Zhixiang
Xu, Shijie
Yan, Kaicheng
Cai, Xuyue
Zhang, Chong
Li, Yulong
Stefanidis, Angelos
Nguyen, Anh
Su, Jionglong
author_facet Lu, Zhixiang
Xu, Shijie
Yan, Kaicheng
Cai, Xuyue
Zhang, Chong
Li, Yulong
Stefanidis, Angelos
Nguyen, Anh
Su, Jionglong
contents The deployment of vision-language models (VLMs) in dermatology is hindered by the trilemma of high computational costs, extreme data scarcity, and the black-box nature of deep learning. To address these challenges, we present SkinCLIP-VL, a resource-efficient framework that adapts foundation models for trustworthy skin cancer diagnosis. Adopting a frozen perception, adaptive reasoning paradigm, we integrate a frozen CLIP encoder with a lightweight, quantized Qwen2.5-VL via low-rank adaptation (LoRA). To strictly align visual regions with clinical semantics under long-tailed distributions, we propose the Consistency-aware Focal Alignment (CFA) Loss. This objective synergizes focal re-weighting, semantic alignment, and calibration. On ISIC and Derm7pt benchmarks, SkinCLIP-VL surpasses 13B-parameter baselines by 4.3-6.2% in accuracy with 43% fewer parameters. Crucially, blinded expert evaluation and out-of-distribution testing confirm that our visually grounded rationales significantly enhance clinical trust compared to traditional saliency maps.
format Preprint
id arxiv_https___arxiv_org_abs_2603_21010
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle SkinCLIP-VL: Consistency-Aware Vision-Language Learning for Multimodal Skin Cancer Diagnosis
Lu, Zhixiang
Xu, Shijie
Yan, Kaicheng
Cai, Xuyue
Zhang, Chong
Li, Yulong
Stefanidis, Angelos
Nguyen, Anh
Su, Jionglong
Computer Vision and Pattern Recognition
The deployment of vision-language models (VLMs) in dermatology is hindered by the trilemma of high computational costs, extreme data scarcity, and the black-box nature of deep learning. To address these challenges, we present SkinCLIP-VL, a resource-efficient framework that adapts foundation models for trustworthy skin cancer diagnosis. Adopting a frozen perception, adaptive reasoning paradigm, we integrate a frozen CLIP encoder with a lightweight, quantized Qwen2.5-VL via low-rank adaptation (LoRA). To strictly align visual regions with clinical semantics under long-tailed distributions, we propose the Consistency-aware Focal Alignment (CFA) Loss. This objective synergizes focal re-weighting, semantic alignment, and calibration. On ISIC and Derm7pt benchmarks, SkinCLIP-VL surpasses 13B-parameter baselines by 4.3-6.2% in accuracy with 43% fewer parameters. Crucially, blinded expert evaluation and out-of-distribution testing confirm that our visually grounded rationales significantly enhance clinical trust compared to traditional saliency maps.
title SkinCLIP-VL: Consistency-Aware Vision-Language Learning for Multimodal Skin Cancer Diagnosis
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.21010