Selected work

Representative Papers

Mixed Magnification Aggregation for Generalizable Region-Level Representations in Computational Pathology

Feb 25, 2026

This work addresses the limitations of existing computational pathology methods, which predominantly rely on single 20× magnification image patches and struggle to effectively model multiscale tissue features and spatial context. To overcome this, the authors propose a region-level hybrid-magnification encoder that systematically explores fusion mechanisms across different magnification levels for the first time. By aggregating information at the region level, the method constructs efficient representations while incorporating a self-supervised pretraining strategy based on masked embedding modeling to balance multiscale contextual awareness with computational efficiency. Evaluated on biomarker prediction tasks across multiple cancer types, the approach demonstrates significant performance gains, underscoring the critical importance of multiscale spatial context in pathological analysis.

0 citationsRead paper

PRISM2: Unlocking Multi-Modal General Pathology AI with Clinical Dialogue

Jun 16, 2025

Current pathology foundation models lack whole-slide image (WSI) comprehension capability and are not trained on large-scale real-world clinical data, limiting their generalizability and clinical utility. This work introduces the first multimodal WSI foundation model designed explicitly for clinical diagnosis. We propose a novel two-stage, clinical-dialogue-driven training paradigm, leveraging 2.3 million WSIs paired with authentic pathology reports to achieve robust vision–language alignment. Our method integrates contrastive learning with image-captioning joint pretraining, coupled with a phased optimization strategy involving freezing and unfreezing of the language model. The resulting model enables zero-shot yes/no classification without prompt engineering or explicit class enumeration. It significantly outperforms baselines—including PRISM and TITAN—on diagnostic classification and biomarker prediction tasks, and surpasses CLIP-based approaches in zero-shot yes/no classification.

0 citationsRead paper

Recent publications

Latest Papers

Mixed Magnification Aggregation for Generalizable Region-Level Representations in Computational Pathology

Feb 25, 2026

This work addresses the limitations of existing computational pathology methods, which predominantly rely on single 20× magnification image patches and struggle to effectively model multiscale tissue features and spatial context. To overcome this, the authors propose a region-level hybrid-magnification encoder that systematically explores fusion mechanisms across different magnification levels for the first time. By aggregating information at the region level, the method constructs efficient representations while incorporating a self-supervised pretraining strategy based on masked embedding modeling to balance multiscale contextual awareness with computational efficiency. Evaluated on biomarker prediction tasks across multiple cancer types, the approach demonstrates significant performance gains, underscoring the critical importance of multiscale spatial context in pathological analysis.

0 citationsRead paper

PRISM2: Unlocking Multi-Modal General Pathology AI with Clinical Dialogue

Jun 16, 2025

Current pathology foundation models lack whole-slide image (WSI) comprehension capability and are not trained on large-scale real-world clinical data, limiting their generalizability and clinical utility. This work introduces the first multimodal WSI foundation model designed explicitly for clinical diagnosis. We propose a novel two-stage, clinical-dialogue-driven training paradigm, leveraging 2.3 million WSIs paired with authentic pathology reports to achieve robust vision–language alignment. Our method integrates contrastive learning with image-captioning joint pretraining, coupled with a phased optimization strategy involving freezing and unfreezing of the language model. The resulting model enables zero-shot yes/no classification without prompt engineering or explicit class enumeration. It significantly outperforms baselines—including PRISM and TITAN—on diagnostic classification and biomarker prediction tasks, and surpasses CLIP-based approaches in zero-shot yes/no classification.

0 citationsRead paper