LEAF-Mamba: Feature Adaptation and Fusion for Whole Slide Image Analysis
Motivated by these challenges, we introduce LEAF-Mamba, a framework that jointly addresses latent evidence adaptation and fusion for frozen-feature WSI analysis.
ProofPaper ↗
Key points
- Whole slide image (WSI) analysis enables computational diagnosis and prognosis, but gigapixel images and limited annotations make slide-level learning challenging.
- Multiple instance learning addresses this task by aggregating patch representations from frozen pretrained encoders.
- Recently, Mamba has enabled efficient modeling of long patch sequences.
- Parallel Mamba and convolutional streams then capture long-range dependencies and sequence-local interactions, with an input-dependent gate modulating their joint representation before slide-level pooling.
Sources (1)
- [1]LEAF-Mamba: Feature Adaptation and Fusion for Whole Slide Image AnalysisarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 21, 01:46 PM
Motivated by these challenges, we introduce LEAF-Mamba, a framework that jointly addresses latent evidence adaptation and fusion for frozen-feature WSI analysis.
Whole slide image (WSI) analysis enables computational diagnosis and prognosis, but gigapixel images and limited annotations make slide-level learning challenging.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 21, 2026PACE: Periodic and Channel-Aligned Encoding of Four-Bit Recurrent States
- Oct 20, 2026SSMCGM-Stream: A Streaming State-Space Model for Personalized Diabetes Management
- Oct 20, 2026GraphCast-Mamba: Learning Non-Markovian Corrections for Weather Forecasting
- Oct 20, 2026Model- and Layer-Level Closed-Loop Forecasting with SSM-Based Spiking Neural Networks
- Oct 20, 2026AGGR-Mamba: Mamba-based Network with Attention-Guided Gated Residual Refinement for Undersampled MRI Reconstruction
- Oct 19, 2026Evaluating Agents on Highly Subjective Tasks