SOURCE-LINKED INTELLIGENCE

SLICEChat: Progressive In-Encoder Token Pruning for Whole-Slide Pathology Language Models

arXiv · AI, language, vision and robotics · article · Sep 21, 2026 · UTC

Whole-slide pathology images (WSIs) contain gigapixel-scale visual content, creating a major scalability challenge for slide-level multimodal large language models (MLLMs). Existing approaches process thousands of patch tokens and typically apply compression only after slide encoding, leaving multimodal attention computationally expensive. We introduce SLICEChat, a slide-level MLLM that integrates progressive token pruning within a hybrid Mamba--Transformer slide encoder. Mamba layers enable efficient long-range propagation, while Transformer layers preserve global interactions as the sequence

Read original source ↗ Open in workspace

recordType

paper

region

Global