TC-SSA: Token Compression via Semantic Slot Aggregation for Gigapixel Pathology Reasoning

• URL: http://arxiv.org/abs/2603.01143v1
• Date: Sun, 01 Mar 2026 15:07:57 GMT
• Title: TC-SSA: Token Compression via Semantic Slot Aggregation for Gigapixel Pathology Reasoning
• Abstract summary: We propose TC-SSA, a learnable token compression framework that aggregates patch features into a fixed number of semantic slots.<n>A gated routing module assigns patches to slots using sparse Top-2 routing, followed by weighted aggregation, enabling global slide coverage under a strict token budget.<n>On the SlideBench(TCGA), our model achieves 78.34% overall accuracy and 77.14% on the diagnosis subset, outperforming sampling-based baselines.
• Score: 10.675615956951468
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: The application of large vision-language models to computational pathology holds great promise for diagnostic assistants but faces a critical computational bottleneck: the gigapixel scale of Whole Slide Images (WSIs). A single WSI typically contains over 105 patches, creating sequence lengths that exceed the constraints of standard Transformer architectures. Existing solutions often resort to spatial sampling, which risks discarding diagnostically critical evidence. To address this, we propose TC-SSA (Token Compression via Semantic Slot Aggregation), a learnable token compression framework that aggregates patch features into a fixed number of semantic slots. A gated routing module assigns patches to slots using sparse Top-2 routing, followed by weighted aggregation, enabling global slide coverage under a strict token budget. The resulting representation retains diagnostically relevant information while reducing the number of visual tokens to 1.7% of the original sequence. On the SlideBench(TCGA), our model achieves 78.34% overall accuracy and 77.14% on the diagnosis subset, outperforming sampling-based baselines under comparable token budgets. The method also generalizes to MIL classification, reaching AUC of 95.83% on TCGA-BRCA, 98.27% on TCGA-NSCLC and 79.80% on PANDA. These results suggest that learnable semantic aggregation provides an effective trade-off between efficiency and diagnostic performance for gigapixel pathology reasoning.

Related papers

• Learnable Token Sparsification for Efficient Gigapixel Whole Slide Image Reasoning [8.596353244934853]
  We reformulate token reduction in whole slide images as a trainable sparsification problem.<n>We show that our framework achieves 73.32% overall accuracy on SlideBench (TCGA)<n>This work provides a highly efficient paradigm for end to end gigapixel whole slide image reasoning.
  arXiv  Detail & Related papers  (2026-06-07T14:07:28Z)
• Thinking in Scales: Accelerating Gigapixel Pathology Image Analysis via Adaptive Continuous Reasoning [52.41928980786654]
  We propose PathCTM to enable token-efficient scale-space continuous reasoning for gigapixel whole slide images.<n>PathCTM formulates diagnostic inference as a dynamic sequential information pursuit.<n>It reduces the number of required image patches by 95.95% and shortens inference time by approximately 95.62%, while maintaining AUC without degradation.
  arXiv  Detail & Related papers  (2026-05-19T07:46:44Z)
• Prune Redundancy, Preserve Essence: Vision Token Compression in VLMs via Synergistic Importance-Diversity [54.95089105944234]
  Vision-language models (VLMs) face significant computational inefficiencies caused by excessive generation of visual tokens.<n>We propose PruneSID, a training-free Synergistic Importance-Diversity approach featuring a two-stage pipeline.<n>We demonstrate state-of-the-art performance, achieving 96.3% accuracy on LLaVA-1.5 with only 11.1% token retention, and 92.8% accuracy at extreme compression rates (5.6%) on LLaVA-NeXT.
  arXiv  Detail & Related papers  (2026-03-10T10:31:58Z)
• Cancer-Net PCa-MultiSeg: Multimodal Enhancement of Prostate Cancer Lesion Segmentation Using Synthetic Correlated Diffusion Imaging [55.62977326180104]
  Current deep learning approaches for prostate cancer lesion segmentation achieve limited performance.<n>We investigate synthetic correlated diffusion imaging (CDI$s$) as an enhancement to standard diffusion-based protocols.<n>Our results establish validated integration pathways for CDI$s$ as a practical drop-in enhancement for PCa lesion segmentation tasks.
  arXiv  Detail & Related papers  (2025-11-11T04:16:12Z)
• Revisiting Data Challenges of Computational Pathology: A Pack-based Multiple Instance Learning Framework [47.035885218675126]
  Computational pathology (CPath) digitizes pathology slides into whole slide images (WSIs)<n>WSIs possess extremely long sequence lengths (up to 200K), significant length variations (from 200 to 200K), and limited supervision.<n>We propose a pack-based MIL framework to address these challenges.
  arXiv  Detail & Related papers  (2025-09-25T09:05:40Z)
• AHDMIL: Asymmetric Hierarchical Distillation Multi-Instance Learning for Fast and Accurate Whole-Slide Image Classification [51.525891360380285]
  AHDMIL is an Asymmetric Hierarchical Distillation Multi-Instance Learning framework.<n>It eliminates irrelevant patches through a two-step training process.<n>It consistently outperforms previous state-of-the-art methods in both classification performance and inference speed.
  arXiv  Detail & Related papers  (2025-08-07T07:47:16Z)
• Minimal High-Resolution Patches Are Sufficient for Whole Slide Image Representation via Cascaded Dual-Scale Reconstruction [13.897013242536849]
  Whole-slide image (WSI) analysis remains challenging due to gigapixel scale and sparsely distributed diagnostic regions.<n>We propose a Cascaded Dual-Scale Reconstruction framework, demonstrating that only an average of 9 high-resolution patches per WSI are sufficient for robust slide-level representation.
  arXiv  Detail & Related papers  (2025-08-03T08:01:30Z)
• Towards a Benchmark for Colorectal Cancer Segmentation in Endorectal Ultrasound Videos: Dataset and Model Development [59.74920439478643]
  In this paper, we collect and annotated the first benchmark dataset that covers diverse ERUS scenarios. Our ERUS-10K dataset comprises 77 videos and 10,000 high-resolution annotated frames. We introduce a benchmark model for colorectal cancer segmentation, named the Adaptive Sparse-context TRansformer (ASTR)
  arXiv  Detail & Related papers  (2024-08-19T15:04:42Z)
• LESS: Label-efficient Multi-scale Learning for Cytological Whole Slide Image Screening [19.803614403803962]
  We propose a weakly-supervised Label-Efficient WSI Screening method, dubbed LESS, for cytological WSI analysis with only slide-level labels. We provide appropriate supervision by using slide-level labels to improve the learning of patch-level features. It outperforms state-of-the-art MIL methods on pathology WSIs and realizes automatic cytological WSI cancer screening.
  arXiv  Detail & Related papers  (2023-06-06T05:09:20Z)
• Cascaded Cross-Attention Networks for Data-Efficient Whole-Slide Image Classification Using Transformers [0.11219061154635457]
  Whole-Slide Imaging allows for the capturing and digitization of high-resolution images of histological specimen. transformer architecture has been proposed as a possible candidate for effectively leveraging the high-resolution information. We propose a novel cascaded cross-attention network (CCAN) based on the cross-attention mechanism that scales linearly with the number of extracted patches.
  arXiv  Detail & Related papers  (2023-05-11T16:42:24Z)
• Hierarchical Transformer for Survival Prediction Using Multimodality Whole Slide Images and Genomics [63.76637479503006]
  Learning good representation of giga-pixel level whole slide pathology images (WSI) for downstream tasks is critical. This paper proposes a hierarchical-based multimodal transformer framework that learns a hierarchical mapping between pathology images and corresponding genes. Our architecture requires fewer GPU resources compared with benchmark methods while maintaining better WSI representation ability.
  arXiv  Detail & Related papers  (2022-11-29T23:47:56Z)
• ClusTR: Exploring Efficient Self-attention via Clustering for Vision Transformers [70.76313507550684]
  We propose a content-based sparse attention method, as an alternative to dense self-attention. Specifically, we cluster and then aggregate key and value tokens, as a content-based method of reducing the total token count. The resulting clustered-token sequence retains the semantic diversity of the original signal, but can be processed at a lower computational cost.
  arXiv  Detail & Related papers  (2022-08-28T04:18:27Z)
• WSSS4LUAD: Grand Challenge on Weakly-supervised Tissue Semantic Segmentation for Lung Adenocarcinoma [51.50991881342181]
  This challenge includes 10,091 patch-level annotations and over 130 million labeled pixels. First place team achieved mIoU of 0.8413 (tumor: 0.8389, stroma: 0.7931, normal: 0.8919)
  arXiv  Detail & Related papers  (2022-04-13T15:27:05Z)
• Hybrid guiding: A multi-resolution refinement approach for semantic segmentation of gigapixel histopathological images [0.7490318169877296]
  We propose a cascaded convolutional neural network design, called H2G-Net, for semantic segmentation. Design involves a detection stage using a patch-wise method, and a refinement stage using a convolutional autoencoder. Best design achieved a Dice score of 0.933 on an independent test set of 90 WSIs.
  arXiv  Detail & Related papers  (2021-12-07T02:31:29Z)
• An Efficient Cervical Whole Slide Image Analysis Framework Based on Multi-scale Semantic and Spatial Features using Deep Learning [2.7218168309244652]
  This study designs a novel inline connection network (InCNet) by enriching the multi-scale connectivity to build the lightweight model named You Only Look Cytopathology Once (YOLCO) The proposed model allows the input size enlarged to megapixel that can stitch the WSI without any overlap by the average repeats. Based on Transformer for classifying the integrated multi-scale multi-task features, the experimental results appear $0.872$ AUC score better and $2.51times$ faster than the best conventional method in WSI classification.
  arXiv  Detail & Related papers  (2021-06-29T06:24:55Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.