[2505.15504] Exploiting Low-Dimensional Manifold of Features for Few-Shot Whole Slide Image Classification

Computer Science > Computer Vision and Pattern Recognition arXiv:2505.15504 (cs) [Submitted on 21 May 2025 (v1), last revised 2 Mar 2026 (this version, v2)] Title:Exploiting Low-Dimensional Manifold of Features for Few-Shot Whole Slide Image Classification Authors:Conghao Xiong, Zhengrui Guo, Zhe Xu, Yifei Zhang, Raymond Kai-Yu Tong, Si Yong Yeo, Hao Chen, Joseph J. Y. Sung, Irwin King View a PDF of the paper titled Exploiting Low-Dimensional Manifold of Features for Few-Shot Whole Slide Image Classification, by Conghao Xiong and 8 other authors View PDF HTML (experimental) Abstract:Few-shot Whole Slide Image (WSI) classification is severely hampered by overfitting. We argue that this is not merely a data-scarcity issue but a fundamentally geometric problem. Grounded in the manifold hypothesis, our analysis shows that features from pathology foundation models exhibit a low-dimensional manifold geometry that is easily perturbed by downstream models. This insight reveals a key potential issue in downstream multiple instance learning models: linear layers are geometry-agnostic and, as we show empirically, can distort the manifold geometry of the features. To address this, we propose the Manifold Residual (MR) block, a plug-and-play module that is explicitly geometry-aware. The MR block reframes the linear layer as residual learning and decouples it into two pathways: (1) a fixed, random matrix serving as a geometric anchor that approximately preserves topology while also acti...