activity
20242026
collaborators

14 papers

cs.CV2026

NeRD: Neuro-Symbolic Rule Distillation for Efficient Ontology-Grounded Chain-of-Thought in Medical Image Diagnosis

Hongxi Yang, Yiwen Jiang, Siyuan Yan +8

Interpretability is essential for trustworthy medical image diagnosis. However, existing concept-driven interpretable methods have key limitations: Concept Bottleneck Models (CBMs)…

cs.CV2026

DermAgent: A Self-Reflective Agentic System for Dermatological Image Analysis with Multi-Tool Reasoning and Traceable Decision-Making

Yize Liu, Siyuan Yan, Ming Hu +5

Dermatological diagnosis requires integrating fine-grained visual perception with expert clinical knowledge. Although Multimodal Large Language Models (MLLMs) facilitate interactiv…

cs.CV2026

Trustworthy and Fair SkinGPT-R1 for Democratizing Dermatological Reasoning across Diverse Ethnicities

Yuhao Shen, Zhangtianyi Chen, Yuanhao He +14

The clinical translation of dermatological AI is hindered by opaque reasoning and systematic performance disparities across skin tones. Here we present SkinGPT-R1, a multimodal lar…

cs.CV2026

A Vision-Language Foundation Model for Zero-shot Clinical Collaboration and Automated Concept Discovery in Dermatology

Siyuan Yan, Xieji Li, Dan Mo +28

Medical foundation models have shown promise in controlled benchmarks, yet widespread deployment remains hindered by reliance on task-specific fine-tuning. Here, we introduce DermF…

cs.CL2026

DermoGPT: Open Weights and Open Data for Morphology-Grounded Dermatological Reasoning MLLMs

Jinghan Ru, Siyuan Yan, Yuguo Yin +2

Multimodal Large Language Models (MLLMs) show promise for medical applications, yet progress in dermatology lags due to limited training data, narrow task coverage, and lack of cli…

cs.CV2025

Multi-Aspect Knowledge-Enhanced Medical Vision-Language Pretraining with Multi-Agent Data Generation

Xieji Li, Siyuan Yan, Yingsheng Liu +4

Vision-language pretraining (VLP) has emerged as a powerful paradigm in medical image analysis, enabling representation learning from large-scale image-text pairs without relying o…