aiqbench analysis
This research introduces a novel framework for whole-slide image analysis in cancer prognosis that leverages high-level pathology semantics to overcome domain shifts caused by varying staining protocols and scanner hardware. By fusing visual evidence with semantic anchors derived from Visual Question Answering, the model achieves superior zero-shot generalization across unseen clinical centers while explicitly modeling uncertainty. The approach demonstrates that semantic representations offer a more robust alternative to pixel-derived features for building reliable, cross-institutional AI diagnostic tools.
Semantic-Anchored Evidential Fusion for Domain-Robust Whole-Slide Survival Analysis
arXiv:2606.19966v2 Announce Type: replace-cross Abstract: Whole-slide images (WSIs) are widely used for computational cancer prognosis. However, most existing methods primarily focus on in-domain performance and fail to generalize across clinical centers. This limitation stems from their reliance on pixel-derived representations that are highly susceptible to domain-specific artifacts caused by staining protocols and scanner hardware. We hypothesize that high-level pathology semantics, such as tumor grade and micro-environmental architecture, provide a domain-invariant semantic representation that mirrors the robust diagnostic logic of human pathologists. Therefore, we propose a Semantic-Anchored Evidential Fusion Survival (SAEFS) framework, where SAEFS derives semantic anchors from WSIs via Visual Question Answering (VQA), employs a dual-stream WSI evidence extraction architecture, uses Dirichlet-based Subjective Logic to model uncertainty, and fuses semantic and visual evidence through a cautious conjunction rule to avoid overconfident fusion from correlated sources. Trained exclusively on one source domain and evaluated zero-shot across four unseen domains, SAEFS consistently outperforms state-of-the-art models both in prediction accuracy and reliability, improving the average C-index by 10.2%. Quantitative analyses further show that VQA-derived semantic features exhibit significantly lower cross-center divergence than pixel-derived features, highlighting their robustness for cross-center clinical applications.
Abstract:Whole-slide images (WSIs) are widely used for computational cancer prognosis. However, most existing methods primarily focus on in-domain performance and fail to generalize across clinical centers. This limitation stems from their reliance on pixel-derived representations that are highly susceptible to domain-specific artifacts caused by staining protocols and scanner hardware. We hypothesize that high-level pathology semantics, such as tumor grade and micro-environmental architecture, provide a domain-invariant semantic representation that mirrors the robust diagnostic logic of human pathologists. Therefore, we propose a Semantic-Anchored Evidential Fusion Survival (SAEFS) framework, where SAEFS derives semantic anchors from WSIs via Visual Question Answering (VQA), employs a dual-stream WSI evidence extraction architecture, uses Dirichlet-based Subjective Logic to model uncertainty, and fuses semantic and visual evidence through a cautious conjunction rule to avoid overconfident fusion from correlated sources. Trained exclusively on one source domain and evaluated zero-shot across four unseen domains, SAEFS consistently outperforms state-of-the-art models both in prediction accuracy and reliability, improving the average C-index by 10.2%. Quantitative analyses further show that VQA-derived semantic features exhibit significantly lower cross-center divergence than pixel-derived features, highlighting their robustness for cross-center clinical applications.
Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
arXiv:2606.19966 [cs.CV]
(or arXiv:2606.19966v2 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2606.19966 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 281 words
Related coverage
DaX: Learning General Pathology Representations Across Scales
arXiv · Jun 8, 2026
Game Design Creativity evals
LessWrong · Oct 12, 2026
Semantic Context-aware mOdality fUsion Transformer (SCOUT): A Context-Aware Multimodal Transformer for Concept-Grounded Pathology Report Generation
arXiv · Sep 9, 2026
Presentation: Engineering AI for Creativity and Curiosity on Mobile
InfoQ · Jul 21, 2026