dorsal/arxiv
View SchemaSkinFlow: Efficient Information Transmission for Open Dermatological Diagnosis via Dynamic Visual Encoding and Staged RL
| Authors | Lijun Liu, Linwei Chen, Zhishou Zhang, Meng Tian, Hengfu Cui, Ruiyang Li, Zhaocheng Liu, Qiang Ju, Qianxi Li, Hong-Yu Zhou |
|---|---|
| Categories | |
| ArXiv ID | 2601.09136vv1 |
| URL | https://arxiv.org/abs/2601.09136 |
| License | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
Abstract
General-purpose Large Vision-Language Models (LVLMs), despite their massive scale, often falter in dermatology due to "diffuse attention" - the inability to disentangle subtle pathological lesions from background noise. In this paper, we challenge the assumption that parameter scaling is the only path to medical precision. We introduce SkinFlow, a framework that treats diagnosis as an optimization of visual information transmission efficiency. Our approach utilizes a Virtual-Width Dynamic Vision Encoder (DVE) to "unfold" complex pathological manifolds without physical parameter expansion, coupled with a two-stage Reinforcement Learning strategy. This strategy sequentially aligns explicit medical descriptions (Stage I) and reconstructs implicit diagnostic textures (Stage II) within a constrained semantic space. Furthermore, we propose a clinically grounded evaluation protocol that prioritizes diagnostic safety and hierarchical relevance over rigid label matching. Empirical results are compelling: our 7B model establishes a new state-of-the-art on the Fitzpatrick17k benchmark, achieving a +12.06% gain in Top-1 accuracy and a +28.57% boost in Top-6 accuracy over the massive general-purpose models (e.g., Qwen3VL-235B and GPT-5.2). These findings demonstrate that optimizing geometric capacity and information flow yields superior diagnostic reasoning compared to raw parameter scaling.
{
"annotation_id": "5db54463-38ed-4005-9355-cf78c6e19511",
"date_created": "2026-02-17T05:53:20.066000Z",
"date_modified": "2026-02-17T05:53:20.066000Z",
"file_hash": "26ca2a0b4040fdf73f6ce443dc6cc028f0082a1e1aba13cba778379b9c697d00",
"private": false,
"record": {
"abstract": "General-purpose Large Vision-Language Models (LVLMs), despite their massive scale, often falter in dermatology due to \"diffuse attention\" - the inability to disentangle subtle pathological lesions from background noise. In this paper, we challenge the assumption that parameter scaling is the only path to medical precision. We introduce SkinFlow, a framework that treats diagnosis as an optimization of visual information transmission efficiency. Our approach utilizes a Virtual-Width Dynamic Vision Encoder (DVE) to \"unfold\" complex pathological manifolds without physical parameter expansion, coupled with a two-stage Reinforcement Learning strategy. This strategy sequentially aligns explicit medical descriptions (Stage I) and reconstructs implicit diagnostic textures (Stage II) within a constrained semantic space. Furthermore, we propose a clinically grounded evaluation protocol that prioritizes diagnostic safety and hierarchical relevance over rigid label matching. Empirical results are compelling: our 7B model establishes a new state-of-the-art on the Fitzpatrick17k benchmark, achieving a +12.06% gain in Top-1 accuracy and a +28.57% boost in Top-6 accuracy over the massive general-purpose models (e.g., Qwen3VL-235B and GPT-5.2). These findings demonstrate that optimizing geometric capacity and information flow yields superior diagnostic reasoning compared to raw parameter scaling.",
"arxiv_id": "2601.09136",
"authors": [
"Lijun Liu",
"Linwei Chen",
"Zhishou Zhang",
"Meng Tian",
"Hengfu Cui",
"Ruiyang Li",
"Zhaocheng Liu",
"Qiang Ju",
"Qianxi Li",
"Hong-Yu Zhou"
],
"categories": [
"cs.CV",
"cs.AI"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"title": "SkinFlow: Efficient Information Transmission for Open Dermatological Diagnosis via Dynamic Visual Encoding and Staged RL",
"url": "https://arxiv.org/abs/2601.09136",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "f4f82705-15a4-4e00-9d93-24bedd1b8c03",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}