FedSSMCoOp framework enables lightweight federated learning for biomedical image classification

Researchers have introduced FedSSMCoOp, a novel federated learning framework designed for few-shot image classification, particularly in biomedical applications where data privacy is paramount. This framework leverages SSM-based Vision Mamba and Cross Mamba blocks to enable multimodal learning without requiring an external Large Language Model for feature alignment. By optimizing only the soft-prompt and communication-prompt updates, FedSSMCoOp achieves a 1.96 times lighter footprint compared to existing baselines while maintaining stable performance across various biomedical image datasets. AI

IMPACT This framework could enable more efficient and privacy-preserving AI model training in specialized domains with limited data.

RANK_REASON The cluster contains a research paper detailing a new framework for federated learning. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• arXiv
• Cross Mamba
• FedSSMCoOp
• Vision Mamba

• paper
• model release
• infra

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →