paper-with-me

Papers

Make A Long Image Short: Adaptive Token Length for Vision Transformers

2023-07-05 · Qiqi Zhou, Yichen Zhu

The vision transformer is a model that breaks down each image into a sequence of tokens with a fixed length and processes them similarly to words in natural language processing. Although increasing the number of tokens typically results in better performance, it also leads to a considerable increase in computational cost. Motivated by the saying "A picture is worth a thousand words," we propose an innovative approach to accelerate the ViT model by shortening long images. Specifically, we introduce a method for adaptively assigning token length for each image at test time to accelerate inference speed. First, we train a Resizable-ViT (ReViT) model capable of processing input with diverse token lengths. Next, we extract token-length labels from ReViT that indicate the minimum number of tokens required to achieve accurate predictions. We then use these labels to train a lightweight Token-Length Assigner (TLA) that allocates the optimal token length for each image during inference. The TLA enables ReViT to process images with the minimum sufficient number of tokens, reducing token numbers in the ViT model and improving inference speed. Our approach is general and compatible with modern vision transformer architectures, significantly reducing computational costs. We verified the effectiveness of our methods on multiple representative ViT models on image classification and action recognition.

📄 PDF Abstract BibTeX arXiv:2307.02092

Code (0)

등록된 구현이 없습니다.

Tasks

Action Recognitionimage-classificationImage Classification

Methods 이 논문이 사용한 방법론

TLA Please enter a description about the method here
Multi-Head Attention 설명 없음
Attention 설명 없음
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Residual Connection 설명 없음
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…

Similar Papers 제목 키워드 기반

Make A Long Image Short: Adaptive Token Length for Vision Transformers

2021-12-03 · Yichen Zhu, Yuqin Zhu, Jie Du, Yi Wang 외

The vision transformer splits each image into a sequence of tokens with fixed length and processes the tokens in the same way as words in natural language processing. More tokens normally lead to better performance but c…

Action Recognitionimage-classificationImage Classification

Adaptive Semiparametric Language Models

2021-02-04 · Dani Yogatama, Cyprien de Masson d'Autume, Lingpeng Kong

We present a language model that combines a large parametric neural network (i.e., a transformer) with a non-parametric episodic memory component in an integrated architecture. Our model uses extended short-term context …

Language ModelingLanguage Modelling

Learning When to Think: Adaptive Reasoning for Test-Time Compute Allocation

2026-08-20 · Gijs Kassenaar, Zhao Yang, Vincent François-Lavet arxiv

Reasoning language models trained with reinforcement learning typically operate under a fixed token budget rather than an explicitly adaptive one, which can lead to over-computation on easy problems and insufficient comp…

Reinforcement Learning

Selective Structured State-Spaces for Long-Form Video Understanding

2023-03-25 · CVPR 2023 1 · Jue Wang, Wentao Zhu, Pichao Wang, Xiang Yu 외

Effective modeling of complex spatiotemporal dependencies in long-form videos remains an open problem. The recently proposed Structured State-Space Sequence (S4) model with its linear complexity offers a promising direct…

Contrastive LearningFormToken ReductionVideo Classification+1

TokensGen: Harnessing Condensed Tokens for Long Video Generation

2025-07-21 · Wenqi Ouyang, Zeqi Xiao, Danni Yang, Yifan Zhou 외 arxiv

Generating consistent long videos is a complex challenge: while diffusion-based generative models generate visually impressive short clips, extending them to longer durations often leads to memory bottlenecks and long-te…

Video Generation