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

2026-09-21

AI Summary

为解决全切片病理图像处理中的可扩展性问题，SLICEChat采用渐进式编码器内标记修剪方法，有效减少计算成本同时保持高精度。

Abstract

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 is progressively shortened. Between stages, language-supervised, region-aware pruning removes spatially coherent low-utility regions under a controlled keep-rate schedule, producing compact slide representations before multimodal fusion. On SlideBench VQA, SLICEChat achieves 79.84% accuracy on TCGA and 59.09% on BCNB cohorts, outperforming prior slide-level pathology MLLMs, and achieves the highest overall WSI-Bench metrics. It also provides competitive memory usage and the inference latency among the evaluated models. These results demonstrate accurate and computationally efficient multimodal reasoning over gigapixel WSIs.

Problem

• Whole-slide pathology images
• Scalability challenge
• Multimodal large language models
• Token pruning

Innovation

• Progressive Token Pruning
• Hybrid Mamba-Transformer Encoder
• Region-aware Pruning
• Multimodal Fusion