ToaSt: Token Channel Selection and Structured Pruning for Efficient ViT

• URL: http://arxiv.org/abs/2602.15720v2
• Date: Wed, 18 Feb 2026 07:44:58 GMT
• Title: ToaSt: Token Channel Selection and Structured Pruning for Efficient ViT
• Authors: Hyunchan Moon, Cheonjun Park, Steven L. Waslander,
• Abstract summary: Vision Transformers (ViTs) have achieved remarkable success across various vision tasks, yet their deployment is often hindered by prohibitive computational costs.<n>We propose ToaSt, a decoupled framework applying specialized strategies to distinct ViT components.
• Score: 14.21482208417138
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Vision Transformers (ViTs) have achieved remarkable success across various vision tasks, yet their deployment is often hindered by prohibitive computational costs. While structured weight pruning and token compression have emerged as promising solutions, they suffer from prolonged retraining times and global propagation that creates optimization challenges, respectively. We propose ToaSt, a decoupled framework applying specialized strategies to distinct ViT components. We apply coupled head-wise structured pruning to Multi-Head Self-Attention modules, leveraging attention operation characteristics to enhance robustness. For Feed-Forward Networks (over 60\% of FLOPs), we introduce Token Channel Selection (TCS) that enhances compression ratios while avoiding global propagation issues. Our analysis reveals TCS effectively filters redundant noise during selection. Extensive evaluations across nine diverse models, including DeiT, ViT-MAE, and Swin Transformer, demonstrate that ToaSt achieves superior trade-offs between accuracy and efficiency, consistently outperforming existing baselines. On ViT-MAE-Huge, ToaSt achieves 88.52\% accuracy (+1.64 \%) with 39.4\% FLOPs reduction. ToaSt transfers effectively to downstream tasks, achieving 52.2 versus 51.9 mAP on COCO object detection. Code and models will be released upon acceptance.

Related papers

• EdgeFlex-Transformer: Transformer Inference for Edge Devices [2.1130318406254074]
  We propose a lightweight yet effective multi-stage optimization pipeline designed to compress and accelerate Vision Transformers (ViTs)<n>Our methodology combines activation profiling, memory-aware pruning, selective mixed-precision execution, and activation-aware quantization (AWQ) to reduce the model's memory footprint without requiring costly retraining or task-specific fine-tuning.<n>Experiments on CIFAR-10 demonstrate that the fully optimized model achieves a 76% reduction in peak memory usage and over 6x lower latency, while retaining or even improving accuracy compared to the original FP32 baseline.
  arXiv  Detail & Related papers  (2025-12-17T21:45:12Z)
• Towards Stabilized and Efficient Diffusion Transformers through Long-Skip-Connections with Spectral Constraints [51.83081671798784]
  Diffusion Transformers (DiT) have emerged as a powerful architecture for image and video generation, offering superior quality and scalability.<n>DiT's practical application suffers from inherent dynamic feature instability, leading to error amplification during cached inference.<n>We propose Skip-DiT, an image and video generative DiT variant enhanced with Long-Skip-Connections (LSCs) - the key efficiency component in U-Nets.
  arXiv  Detail & Related papers  (2024-11-26T17:28:10Z)
• Dynamic Tuning Towards Parameter and Inference Efficiency for ViT Adaptation [67.13876021157887]
  Dynamic Tuning (DyT) is a novel approach to improve both parameter and inference efficiency for ViT adaptation. DyT achieves superior performance compared to existing PEFT methods while evoking only 71% of their FLOPs on the VTAB-1K benchmark.
  arXiv  Detail & Related papers  (2024-03-18T14:05:52Z)
• The Surprising Effectiveness of Skip-Tuning in Diffusion Sampling [78.6155095947769]
  Skip-Tuning is a simple yet surprisingly effective training-free tuning method on the skip connections. Our method can achieve 100% FID improvement for pretrained EDM on ImageNet 64 with only 19 NFEs (1.75) While Skip-Tuning increases the score-matching losses in the pixel space, the losses in the feature space are reduced.
  arXiv  Detail & Related papers  (2024-02-23T08:05:23Z)
• HEViTPose: High-Efficiency Vision Transformer for Human Pose Estimation [3.1690891866882236]
  This paper proposes a High- Efficiency Vision Transformer for Human Pose Estimation (HEViTPose) In HEViTPose, a Cascaded Group Spatial Reduction Multi-Head Attention Module (CGSR-MHA) is proposed, which reduces the computational cost. Comprehensive experiments on two benchmark datasets (MPII and COCO) demonstrate that the small and large HEViTPose models are on par with state-of-the-art models.
  arXiv  Detail & Related papers  (2023-11-22T06:45:16Z)
• Hierarchical Side-Tuning for Vision Transformers [33.536948382414316]
  Fine-tuning pre-trained Vision Transformers (ViTs) has showcased significant promise in enhancing visual recognition tasks. PETL has shown potential for achieving high performance with fewer parameter updates compared to full fine-tuning. This paper introduces Hierarchical Side-Tuning (HST), an innovative PETL method facilitating the transfer of ViT models to diverse downstream tasks.
  arXiv  Detail & Related papers  (2023-10-09T04:16:35Z)
• Deeper Insights into ViTs Robustness towards Common Corruptions [82.79764218627558]
  We investigate how CNN-like architectural designs and CNN-based data augmentation strategies impact on ViTs' robustness towards common corruptions. We demonstrate that overlapping patch embedding and convolutional Feed-Forward Network (FFN) boost performance on robustness. We also introduce a novel conditional method enabling input-varied augmentations from two angles.
  arXiv  Detail & Related papers  (2022-04-26T08:22:34Z)
• SepViT: Separable Vision Transformer [20.403430632658946]
  Vision Transformers often rely on extensive computational costs to achieve high performance, which is burdensome to deploy on resource-constrained devices. We draw lessons from depthwise separable convolution and imitate its ideology to design an efficient Transformer backbone, i.e., Separable Vision Transformer, abbreviated as SepViT. SepViT helps to carry out the local-global information interaction within and among the windows in sequential order via a depthwise separable self-attention.
  arXiv  Detail & Related papers  (2022-03-29T09:20:01Z)
• AdaViT: Adaptive Tokens for Efficient Vision Transformer [91.88404546243113]
  We introduce AdaViT, a method that adaptively adjusts the inference cost of vision transformer (ViT) for images of different complexity. AdaViT achieves this by automatically reducing the number of tokens in vision transformers that are processed in the network as inference proceeds.
  arXiv  Detail & Related papers  (2021-12-14T18:56:07Z)
• A Unified Pruning Framework for Vision Transformers [40.7622551128182]
  Vision transformer (ViT) and its variants have achieved promising performances in various computer vision tasks. We propose a unified framework for structural pruning of both ViTs and its variants, namely UP-ViTs. Our method focuses on pruning all ViTs components while maintaining the consistency of the model structure.
  arXiv  Detail & Related papers  (2021-11-30T05:01:02Z)
• Global Vision Transformer Pruning with Hessian-Aware Saliency [93.33895899995224]
  This work challenges the common design philosophy of the Vision Transformer (ViT) model with uniform dimension across all the stacked blocks in a model stage. We derive a novel Hessian-based structural pruning criteria comparable across all layers and structures, with latency-aware regularization for direct latency reduction. Performing iterative pruning on the DeiT-Base model leads to a new architecture family called NViT (Novel ViT), with a novel parameter that utilizes parameters more efficiently.
  arXiv  Detail & Related papers  (2021-10-10T18:04:59Z)

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.