Selected work

Representative Papers

Unsupervised Anomaly Detection for Image Dataset Quality Assurance in Multi-Center Breast MRI

Aug 17, 2026

This study addresses quality control challenges in multi-center breast MRI by proposing an unsupervised anomaly detection framework. Through a controlled benchmark encompassing 17 anomaly types, we identify near out-of-distribution (OOD) detection as a critical bottleneck. To overcome this, we introduce a novel approach integrating projection methods, 3D reconstruction, and hybrid OOD detection, enhanced with domain-specific features and positional encoding for precise identification. Experiments demonstrate that the projection method achieves an AUROC of 0.954, while 3D reconstruction exhibits superior generalization, underscoring the necessity of domain adaptation. This research establishes a scalable, automated quality control paradigm, providing essential benchmarks and methodological support for data safety in medical AI, although detecting implants remains challenging.

0 citationsRead paper

Cross-Modal-Domain Generalization Through Semantically Aligned Discrete Representations

May 12, 2026

This work addresses the challenge of simultaneously achieving cross-modal generalization and preserving modality-specific characteristics in multimodal representation learning. To this end, we propose CoDAAR, a novel framework that constructs the first competition-free unified discrete representation space. CoDAAR leverages Discrete Temporal Alignment (DTA) and Cascaded Semantic Alignment (CSA) mechanisms to establish cross-modal semantic consensus while retaining modality uniqueness. Trained via a self-supervised reconstruction objective, the method overcomes inherent limitations of both continuous and discrete representation approaches. Extensive experiments demonstrate that CoDAAR achieves state-of-the-art performance across diverse tasks—including event classification, temporal localization, video segmentation, and cross-dataset transfer—establishing a new discrete paradigm for multimodal representation learning.

0 citationsRead paper

Sheaf Diffusion with Adaptive Local Structure for Spatio-Temporal Forecasting

Apr 13, 2026

Traditional message-passing approaches struggle to capture high-order interactions arising from local heterogeneity in spatiotemporal systems, thereby limiting predictive performance. This work reframes spatiotemporal forecasting as a problem of learning information flow within locally structured spaces and proposes a sheaf-based diffusion graph neural network. It introduces, for the first time, dynamic, locally adaptive, learnable linear restriction maps that explicitly model latent local structures. The method effectively mitigates the oversmoothing issue prevalent in deep GNNs and enhances model expressiveness. Extensive experiments on multiple real-world spatiotemporal forecasting benchmarks demonstrate state-of-the-art performance, underscoring the strong potential of sheaf-theoretic topological representations for spatiotemporal graph learning.

0 citationsRead paper

Explainable histomorphology-based survival prediction of glioblastoma, IDH-wildtype

Jan 16, 2026

This study addresses the challenge of automatically extracting interpretable prognostic morphological features from whole-slide images of IDH wild-type glioblastoma to predict patient survival. We propose a novel interpretable multiple instance learning framework that, for the first time, integrates a sparse autoencoder with a Cox proportional hazards model. Evaluated on 720 real-world cases, the model achieves an AUC of 0.67 in survival stratification. It identifies 24 visual patches significantly associated with survival, 21 of which were validated by neuropathologists and categorized into seven distinct histological feature classes. This approach enables an automatic mapping from raw histopathology images to clinically comprehensible morphological patterns, substantially enhancing model transparency and pathological relevance.

0 citationsRead paper

What Is Novel? A Knowledge-Driven Framework for Bias-Aware Literature Originality Evaluation

Jan 14, 2026

Current approaches to assessing the novelty of scientific papers rely heavily on subjective judgment and lack systematic, interpretable, and reviewer-aligned objective methods. This work proposes the first knowledge-driven framework for novelty evaluation, explicitly modeling human judgments of novelty derived from peer review comments across nearly 80,000 top-tier AI conference papers. By integrating structured paper representations with a semantic similarity graph of related literature, the framework enables fine-grained, concept-level originality comparisons. It combines large language model fine-tuning, knowledge extraction, and semantic retrieval to produce calibrated, interpretable novelty scores that significantly outperform existing methods in accuracy, consistency, and alignment with human reviewers.

0 citationsRead paper

Recent publications

Latest Papers

Unsupervised Anomaly Detection for Image Dataset Quality Assurance in Multi-Center Breast MRI

Aug 17, 2026

This study addresses quality control challenges in multi-center breast MRI by proposing an unsupervised anomaly detection framework. Through a controlled benchmark encompassing 17 anomaly types, we identify near out-of-distribution (OOD) detection as a critical bottleneck. To overcome this, we introduce a novel approach integrating projection methods, 3D reconstruction, and hybrid OOD detection, enhanced with domain-specific features and positional encoding for precise identification. Experiments demonstrate that the projection method achieves an AUROC of 0.954, while 3D reconstruction exhibits superior generalization, underscoring the necessity of domain adaptation. This research establishes a scalable, automated quality control paradigm, providing essential benchmarks and methodological support for data safety in medical AI, although detecting implants remains challenging.

0 citationsRead paper

Cross-Modal-Domain Generalization Through Semantically Aligned Discrete Representations

May 12, 2026

This work addresses the challenge of simultaneously achieving cross-modal generalization and preserving modality-specific characteristics in multimodal representation learning. To this end, we propose CoDAAR, a novel framework that constructs the first competition-free unified discrete representation space. CoDAAR leverages Discrete Temporal Alignment (DTA) and Cascaded Semantic Alignment (CSA) mechanisms to establish cross-modal semantic consensus while retaining modality uniqueness. Trained via a self-supervised reconstruction objective, the method overcomes inherent limitations of both continuous and discrete representation approaches. Extensive experiments demonstrate that CoDAAR achieves state-of-the-art performance across diverse tasks—including event classification, temporal localization, video segmentation, and cross-dataset transfer—establishing a new discrete paradigm for multimodal representation learning.

0 citationsRead paper

Sheaf Diffusion with Adaptive Local Structure for Spatio-Temporal Forecasting

Apr 13, 2026

Traditional message-passing approaches struggle to capture high-order interactions arising from local heterogeneity in spatiotemporal systems, thereby limiting predictive performance. This work reframes spatiotemporal forecasting as a problem of learning information flow within locally structured spaces and proposes a sheaf-based diffusion graph neural network. It introduces, for the first time, dynamic, locally adaptive, learnable linear restriction maps that explicitly model latent local structures. The method effectively mitigates the oversmoothing issue prevalent in deep GNNs and enhances model expressiveness. Extensive experiments on multiple real-world spatiotemporal forecasting benchmarks demonstrate state-of-the-art performance, underscoring the strong potential of sheaf-theoretic topological representations for spatiotemporal graph learning.

0 citationsRead paper

Explainable histomorphology-based survival prediction of glioblastoma, IDH-wildtype

Jan 16, 2026

This study addresses the challenge of automatically extracting interpretable prognostic morphological features from whole-slide images of IDH wild-type glioblastoma to predict patient survival. We propose a novel interpretable multiple instance learning framework that, for the first time, integrates a sparse autoencoder with a Cox proportional hazards model. Evaluated on 720 real-world cases, the model achieves an AUC of 0.67 in survival stratification. It identifies 24 visual patches significantly associated with survival, 21 of which were validated by neuropathologists and categorized into seven distinct histological feature classes. This approach enables an automatic mapping from raw histopathology images to clinically comprehensible morphological patterns, substantially enhancing model transparency and pathological relevance.

0 citationsRead paper

What Is Novel? A Knowledge-Driven Framework for Bias-Aware Literature Originality Evaluation

Jan 14, 2026

Current approaches to assessing the novelty of scientific papers rely heavily on subjective judgment and lack systematic, interpretable, and reviewer-aligned objective methods. This work proposes the first knowledge-driven framework for novelty evaluation, explicitly modeling human judgments of novelty derived from peer review comments across nearly 80,000 top-tier AI conference papers. By integrating structured paper representations with a semantic similarity graph of related literature, the framework enables fine-grained, concept-level originality comparisons. It combines large language model fine-tuning, knowledge extraction, and semantic retrieval to produce calibrated, interpretable novelty scores that significantly outperform existing methods in accuracy, consistency, and alignment with human reviewers.

0 citationsRead paper