Evaluating Explanation Methods by the Predictors They Induce
本文提出一种通过重建预测来评估机器学习模型解释方法的新测试，适用于PDP、ALE、SHAP和LIME等，并在不同数据集上验证了其有效性。
0 citationsRead paper
Selected work
Representative Papers
A Smaller Transformer in Your Transformer
该研究通过提出Transformer-Within-Transformer方法，解决了Vision Transformers中的计算冗余问题，减少了参数数量和推理计算量，同时保持了模型性能。
0 citationsRead paper
Finite-Mask Gaussian-Process Reconstruction on a Periodic Sensor Ring: Mask-Geometry Dependence, Fourier-Mode Coupling, and Posterior Trace
研究了周期性传感器环上任意观测掩模对高斯过程重建后验协方差、傅里叶模式耦合和归一化后验迹的影响，通过分析不同掩模几何形状导致的不同效果。
0 citationsRead paper
Deep Learning Segmentation of Diffusion-Weighted MRI Acute Ischaemic Stroke: A Pragmatic Evaluation Across Three Datasets
研究使用深度学习方法对急性缺血性脑卒中进行分割，通过对比不同输入配置和模型架构，发现基于DWI的基线nnU-Net模型无需预处理即可实现快速准确的病灶分割。
0 citationsRead paper
Adaptive Post-Processing Drives Instance-Level Detection in Stroke Lesion Segmentation
This study addresses the misalignment between instance detection and voxel-level metrics, as well as the frequent miss-detection of small lesions in stroke segmentation. We propose a Volume-Conditioned Adaptive Post-processing (VCAP) scheme alongside the Viola2Plus architecture. By employing dynamic threshold adjustment to bridge detection gaps and integrating resolution-aware attention with a dual-architecture ensemble, our method significantly enhances small target detectability. Notably, results demonstrate that optimized post-processing contributes more to instance detection performance than architectural modifications alone. Five-fold cross-validation achieved a Dice score of 0.651 and a Lesion-F1 of 0.614, with a 3.7% improvement in small lesion detection rate, effectively mitigating the limitations of conventional evaluation metrics.
0 citationsRead paper
Recent publications
Latest Papers
Evaluating Explanation Methods by the Predictors They Induce
本文提出一种通过重建预测来评估机器学习模型解释方法的新测试，适用于PDP、ALE、SHAP和LIME等，并在不同数据集上验证了其有效性。
0 citationsRead paper
A Smaller Transformer in Your Transformer
该研究通过提出Transformer-Within-Transformer方法，解决了Vision Transformers中的计算冗余问题，减少了参数数量和推理计算量，同时保持了模型性能。
0 citationsRead paper
Finite-Mask Gaussian-Process Reconstruction on a Periodic Sensor Ring: Mask-Geometry Dependence, Fourier-Mode Coupling, and Posterior Trace
研究了周期性传感器环上任意观测掩模对高斯过程重建后验协方差、傅里叶模式耦合和归一化后验迹的影响，通过分析不同掩模几何形状导致的不同效果。
0 citationsRead paper
Deep Learning Segmentation of Diffusion-Weighted MRI Acute Ischaemic Stroke: A Pragmatic Evaluation Across Three Datasets
研究使用深度学习方法对急性缺血性脑卒中进行分割，通过对比不同输入配置和模型架构，发现基于DWI的基线nnU-Net模型无需预处理即可实现快速准确的病灶分割。
0 citationsRead paper
Adaptive Post-Processing Drives Instance-Level Detection in Stroke Lesion Segmentation
This study addresses the misalignment between instance detection and voxel-level metrics, as well as the frequent miss-detection of small lesions in stroke segmentation. We propose a Volume-Conditioned Adaptive Post-processing (VCAP) scheme alongside the Viola2Plus architecture. By employing dynamic threshold adjustment to bridge detection gaps and integrating resolution-aware attention with a dual-architecture ensemble, our method significantly enhances small target detectability. Notably, results demonstrate that optimized post-processing contributes more to instance detection performance than architectural modifications alone. Five-fold cross-validation achieved a Dice score of 0.651 and a Lesion-F1 of 0.614, with a 3.7% improvement in small lesion detection rate, effectively mitigating the limitations of conventional evaluation metrics.
0 citationsRead paper