paper-with-me

Papers

TinyDrop: Tiny Model Guided Token Dropping for Vision Transformers

2025-09-03 · Guoxin Wang, Qingyuan Wang, Binhua Huang, Shaowu Chen, Deepu John 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 propose TinyDrop, a training-free token dropping framework guided by a lightweight vision model. The guidance model estimates the importance of tokens while performing inference, thereby selectively discarding low-importance tokens if large vit models need to perform attention calculations. The framework operates plug-and-play, requires no architectural modifications, and is compatible with diverse ViT architectures. Evaluations on standard image classification benchmarks demonstrate that our framework reduces FLOPs by up to 80% for ViTs with minimal accuracy degradation, highlighting its generalization capability and practical utility for efficient ViT-based classification.

📄 PDF Abstract BibTeX arXiv:2509.03379

Code (0)

등록된 구현이 없습니다.

Tasks

Image Classification

Similar Papers 제목 키워드 기반

HiRED: Attention-Guided Token Dropping for Efficient Inference of High-Resolution Vision-Language Models

2024-08-20 · Kazi Hasan Ibn Arif, JinYi Yoon, Dimitrios S. Nikolopoulos, Hans Vandierendonck 외

High-resolution Vision-Language Models (VLMs) are widely used in multimodal tasks to enhance accuracy by preserving detailed image information. However, these models often generate an excessive number of visual tokens du…

GPULanguage ModellingLarge Language Model

Training Noise Token Pruning

2024-11-27 · Mingxing Rao, Bohan Jiang, Daniel Moyer

In the present work we present Training Noise Token (TNT) Pruning for vision transformers. Our method relaxes the discrete token dropping condition to continuous additive noise, providing smooth optimization in training,…

Self-slimmed Vision Transformer

2021-11-24 · Zhuofan Zong, Kunchang Li, Guanglu Song, Yali Wang 외

Vision transformers (ViTs) have become the popular structures and outperformed convolutional neural networks (CNNs) on various vision tasks. However, such powerful transformers bring a huge computation burden, because of…

Knowledge Distillation

CacheFlow: Compressive Streaming Memory for Efficient Long-Form Video Understanding

2025-11-17 · Shrenik Patel, Daivik Patel arxiv

Long-form video question answering (VQA) overwhelms current vision-language models (VLMs) because attention and key-value (KV) caches grow with runtime, forcing either expensive inference or near-sighted sliding windows.…

Video Question Answering

Multi-Stage Vision Token Dropping: Towards Efficient Multimodal Large Language Model

2024-11-16 · Ting Liu, Liangtao Shi, Richang Hong, Yue Hu 외

The vision tokens in multimodal large language models usually exhibit significant spatial and temporal redundancy and take up most of the input tokens, which harms their inference efficiency. To solve this problem, some …

Language ModelingLanguage ModellingLarge Language ModelMultimodal Large Language Model