Researchers have developed a new framework called Multi-View Synergistic Learning (MVSL) to improve biomedical image classification in low-resource settings. MVSL decouples the adaptation of visual and textual encoders for more stable parameter-efficient fine-tuning. It also incorporates multi-granularity contrastive learning to better distinguish between visually similar diseases and uses large language models to preserve disease-level semantic structure. Experiments across eleven datasets show MVSL outperforms existing methods in few-shot and zero-shot classification. AI
Summary written by gemini-2.5-flash-lite from 2 sources. How we write summaries →
IMPACT Enhances low-resource biomedical image classification by improving fine-grained discrimination and leveraging LLMs for semantic structure.
RANK_REASON This is a research paper detailing a new framework for biomedical image classification.