paper-with-me

Papers

ToSA: Token Selective Attention for Efficient Vision Transformers

2024-06-13 · Manish Kumar Singh, Rajeev Yasarla, Hong Cai, Mingu Lee, Fatih Porikli

In this paper, we propose a novel token selective attention approach, ToSA, which can identify tokens that need to be attended as well as those that can skip a transformer layer. More specifically, a token selector parses the current attention maps and predicts the attention maps for the next layer, which are then used to select the important tokens that should participate in the attention operation. The remaining tokens simply bypass the next layer and are concatenated with the attended ones to re-form a complete set of tokens. In this way, we reduce the quadratic computation and memory costs as fewer tokens participate in self-attention while maintaining the features for all the image patches throughout the network, which allows it to be used for dense prediction tasks. Our experiments show that by applying ToSA, we can significantly reduce computation costs while maintaining accuracy on the ImageNet classification benchmark. Furthermore, we evaluate on the dense prediction task of monocular depth estimation on NYU Depth V2, and show that we can achieve similar depth prediction accuracy using a considerably lighter backbone with ToSA.

📄 PDF Abstract BibTeX arXiv:2406.08816

Code (0)

등록된 구현이 없습니다.

Tasks

Depth EstimationDepth PredictionMonocular Depth EstimationPrediction

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

ToSA: Token Merging with Spatial Awareness

2025-06-24 · Hsiang-Wei Huang, Wenhao Chai, Kuang-Ming Chen, Cheng-Yen Yang 외

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 merg…

Embodied Question AnsweringQuestion Answering

MambaMixer: Efficient Selective State Space Models with Dual Token and Channel Selection

2024-03-29 · Ali Behrouz, Michele Santacatterina, Ramin Zabih

Recent advances in deep learning have mainly relied on Transformers due to their data dependency and ability to learn at scale. The attention module in these architectures, however, exhibits quadratic time and space in i…

channel selectionImage Classificationobject-detectionObject Detection+4

TinyDrop: Tiny Model Guided Token Dropping for Vision Transformers

2025-09-03 · Guoxin Wang, Qingyuan Wang, Binhua Huang, Shaowu Chen 외 arxiv

Vision Transformers (ViTs) achieve strong performance in image classification but incur high computational costs from processing all image tokens. To reduce inference costs in large ViTs without compromising accuracy, we…

Image Classification

Vision Transformers Need Better Token Interaction

2026-05-22 · Linxiang Su arxiv

Vision Transformers (ViTs) can learn strong image-level representations while their patch representations become less effective for dense prediction during prolonged training. We revisit this dense degradation phenomenon…

Semantic Segmentation

Selective Induction Heads: How Transformers Select Causal Structures In Context

2025-09-09 · Francesco D'Angelo, Francesco Croce, Nicolas Flammarion arxiv

Transformers have exhibited exceptional capabilities in sequence modeling tasks, leveraging self-attention and in-context learning. Critical to this success are induction heads, attention circuits that enable copying tok…