Context Matters: Query-aware Dynamic Long Sequence Modeling of Gigapixel Images

TL;DR

This paper introduces Querent, a query-aware dynamic modeling framework for gigapixel images that approximates full self-attention efficiently, enabling accurate analysis of large pathology images with reduced computational costs.

Contribution

Querent is the first method to adaptively focus on relevant regions for each patch, approximating full self-attention with theoretical bounds and practical efficiency.

Findings

Outperforms state-of-the-art methods on multiple WSI tasks

Reduces computational overhead significantly

Maintains global context for accurate predictions

Abstract

Whole slide image (WSI) analysis presents significant computational challenges due to the massive number of patches in gigapixel images. While transformer architectures excel at modeling long-range correlations through self-attention, their quadratic computational complexity makes them impractical for computational pathology applications. Existing solutions like local-global or linear self-attention reduce computational costs but compromise the strong modeling capabilities of full self-attention. In this work, we propose Querent, i.e., the query-aware long contextual dynamic modeling framework, which achieves a theoretically bounded approximation of full self-attention while delivering practical efficiency. Our method adaptively predicts which surrounding regions are most relevant for each patch, enabling focused yet unrestricted attention computation only with potentially important contexts. By using efficient region-wise metadata computation and importance estimation, our approach dramatically reduces computational overhead while preserving global perception to model fine-grained patch correlations. Through comprehensive experiments on biomarker prediction, gene mutation prediction, cancer subtyping, and survival analysis across over 10 WSI datasets, our method demonstrates superior performance compared to the state-of-the-art approaches. Codes are available at https://github.com/dddavid4real/Querent.