ToSA: Token Merging with Spatial Awareness

• URL: http://arxiv.org/abs/2506.20066v1
• Date: Tue, 24 Jun 2025 23:58:20 GMT
• Title: ToSA: Token Merging with Spatial Awareness
• Authors: Hsiang-Wei Huang, Wenhao Chai, Kuang-Ming Chen, Cheng-Yen Yang, Jenq-Neng Hwang,
• Abstract summary: ToSA is a novel token merging method that combines both semantic and spatial awareness to guide the token merging process.<n>ToSA outperforms previous token merging methods across multiple benchmarks on visual and embodied question answering.
• Score: 20.85218319861827
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Token merging has emerged as an effective strategy to accelerate Vision Transformers (ViT) by reducing computational costs. However, existing methods primarily rely on the visual token's feature similarity for token merging, overlooking the potential of integrating spatial information, which can serve as a reliable criterion for token merging in the early layers of ViT, where the visual tokens only possess weak visual information. In this paper, we propose ToSA, a novel token merging method that combines both semantic and spatial awareness to guide the token merging process. ToSA leverages the depth image as input to generate pseudo spatial tokens, which serve as auxiliary spatial information for the visual token merging process. With the introduced spatial awareness, ToSA achieves a more informed merging strategy that better preserves critical scene structure. Experimental results demonstrate that ToSA outperforms previous token merging methods across multiple benchmarks on visual and embodied question answering while largely reducing the runtime of the ViT, making it an efficient solution for ViT acceleration. The code will be available at: https://github.com/hsiangwei0903/ToSA

Related papers

• Rethinking Visual Token Reduction in LVLMs under Cross-modal Misalignment [38.04426918886084]
  Vision-Language Models (LVLMs) encode visual inputs as dense sequences of patch-level tokens to capture fine-grained semantics.<n>Previous efforts have explored visual token reduction either prior to or within the large language models (LLMs)<n>We introduce VisionDrop, a training-free, visual-only pruning framework that selects informative visual tokens based on intra-modal (visual-to-visual) attention.
  arXiv  Detail & Related papers  (2025-06-27T14:55:40Z)
• Window Token Concatenation for Efficient Visual Large Language Models [59.6094005814282]
  We propose Window Token Concatenation (WiCo) to reduce visual tokens in Visual Large Language Models (VLLMs)<n>WiCo group diverse tokens into one, and thus obscure some fine details.<n>We perform extensive experiments on both coarse- and fine-grained visual understanding tasks based on LLaVA-1.5 and Shikra, showing better performance compared with existing token reduction projectors.
  arXiv  Detail & Related papers  (2025-04-05T02:32:58Z)
• Token Dynamics: Towards Efficient and Dynamic Video Token Representation for Video Large Language Models [50.214593234229255]
  We introduce the novel task of extreme short token reduction, aiming to represent extensive video sequences with a minimal number of tokens.<n>We propose Token Dynamics, a new video representation framework that dynamically reduces token count while preserving spatial-temporal coherence.<n>Experiments demonstrate a reduction of token count to merely 0.07% of the original tokens, with only a minor performance drop of 1.13%.
  arXiv  Detail & Related papers  (2025-03-21T09:46:31Z)
• Learning to Merge Tokens via Decoupled Embedding for Efficient Vision Transformers [18.850145019462552]
  Recent token reduction methods for Vision Transformers (ViTs) incorporate token merging, which measures the similarities between token embeddings and combines the most similar pairs.<n>Our method introduces a lightweight embedding module decoupled from the ViT forward pass to extract dedicated features for token merging.<n>Thanks to the decoupled structure, our method can be seamlessly integrated into existing ViT backbones and trained either modularly by learning only the decoupled embeddings or end-to-end by fine-tuning.
  arXiv  Detail & Related papers  (2024-12-13T21:17:11Z)
• Video Token Merging for Long-form Video Understanding [17.59960070514554]
  We propose a learnable video token merging algorithm that dynamically merges tokens based on their saliency. Our approach significantly reduces memory costs by 84% and boosts throughput by approximately 6.89 times compared to baseline algorithms.
  arXiv  Detail & Related papers  (2024-10-31T09:55:32Z)
• ToSA: Token Selective Attention for Efficient Vision Transformers [50.13756218204456]
  ToSA is a token selective attention approach that can identify tokens that need to be attended as well as those that can skip a transformer layer. We show that ToSA can significantly reduce computation costs while maintaining accuracy on the ImageNet classification benchmark.
  arXiv  Detail & Related papers  (2024-06-13T05:17:21Z)
• LeMeViT: Efficient Vision Transformer with Learnable Meta Tokens for Remote Sensing Image Interpretation [37.72775203647514]
  This paper proposes to use learnable meta tokens to formulate sparse tokens, which effectively learn key information and improve inference speed. By employing Dual Cross-Attention (DCA) in the early stages with dense visual tokens, we obtain the hierarchical architecture LeMeViT with various sizes. Experimental results in classification and dense prediction tasks show that LeMeViT has a significant $1.7 times$ speedup, fewer parameters, and competitive performance compared to the baseline models.
  arXiv  Detail & Related papers  (2024-05-16T03:26:06Z)
• Token Fusion: Bridging the Gap between Token Pruning and Token Merging [71.84591084401458]
  Vision Transformers (ViTs) have emerged as powerful backbones in computer vision, outperforming many traditional CNNs. computational overhead, largely attributed to the self-attention mechanism, makes deployment on resource-constrained edge devices challenging. We introduce "Token Fusion" (ToFu), a method that amalgamates the benefits of both token pruning and token merging.
  arXiv  Detail & Related papers  (2023-12-02T04:29:19Z)
• AiluRus: A Scalable ViT Framework for Dense Prediction [95.1313839257891]
  Vision transformers (ViTs) have emerged as a prevalent architecture for vision tasks owing to their impressive performance. We propose to apply adaptive resolution for different regions in the image according to their importance. We evaluate our proposed method on three different datasets and observe promising performance.
  arXiv  Detail & Related papers  (2023-11-02T12:48:43Z)
• Token-Label Alignment for Vision Transformers [93.58540411138164]
  Data mixing strategies (e.g., CutMix) have shown the ability to greatly improve the performance of convolutional neural networks (CNNs) We identify a token fluctuation phenomenon that has suppressed the potential of data mixing strategies. We propose a token-label alignment (TL-Align) method to trace the correspondence between transformed tokens and the original tokens to maintain a label for each token.
  arXiv  Detail & Related papers  (2022-10-12T17:54:32Z)

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.