# TORE: Token Recycling in Vision Transformers for Efficient Active Visual Exploration

arXiv: https://arxiv.org/abs/2311.15335 · HTML: https://mlsanity.com/paper/2311.15335/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 30 of its 44 references that are in the map.
- First posted on arXiv 26 Nov 2023.

## Builds on (most cited first, 30 of 30)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [Longformer: The Long-Document Transformer](https://mlsanity.com/paper/2004.05150/) (2020)
- [What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?](https://mlsanity.com/paper/1703.04977/) (2017)
- [DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsification](https://mlsanity.com/paper/2106.02034/) (2021)
- [Continual Test-Time Domain Adaptation](https://mlsanity.com/paper/2203.13591/) (2022)
- [Class-incremental learning: survey and performance evaluation on image classification](https://mlsanity.com/paper/2010.15277/) (2020)
- [Transfusion: Understanding Transfer Learning for Medical Imaging](https://mlsanity.com/paper/1902.07208/) (2019)
- [AdaViT: Adaptive Tokens for Efficient Vision Transformer](https://mlsanity.com/paper/2112.07658/) (2021)
- [EfficientFormer: Vision Transformers at MobileNet Speed](https://mlsanity.com/paper/2206.01191/) (2022)
- [Star-Transformer](https://mlsanity.com/paper/1902.09113/) (2019)
- [Sensitivity-Aware Visual Parameter-Efficient Fine-Tuning](https://mlsanity.com/paper/2303.08566/) (2023)
- [An Exploration of Embodied Visual Exploration](https://mlsanity.com/paper/2001.02192/) (2020)
- [Glance and Focus Networks for Dynamic Visual Recognition](https://mlsanity.com/paper/2201.03014/) (2022)
- [Zero Time Waste: Recycling Predictions in Early Exit Neural Networks](https://mlsanity.com/paper/2106.05409/) (2021)
- [Learning to Look Around: Intelligently Exploring Unseen Environments for Unknown Tasks](https://mlsanity.com/paper/1709.00507/) (2017)
- [Test-time Unsupervised Domain Adaptation](https://mlsanity.com/paper/2010.01926/) (2020)
- [Expediting Large-Scale Vision Transformer for Dense Prediction without Fine-tuning](https://mlsanity.com/paper/2210.01035/) (2022)
- [ICICLE: Interpretable Class Incremental Continual Learning](https://mlsanity.com/paper/2303.07811/) (2023)
- [Combining Weakly and Webly Supervised Learning for Classifying Food Images](https://mlsanity.com/paper/1712.08730/) (2017)
- [Active Visual Exploration Based on Attention-Map Entropy](https://mlsanity.com/paper/2303.06457/) (2023)
- [Consistency driven Sequential Transformers Attention Model for Partially Observable Scenes](https://mlsanity.com/paper/2204.00656/) (2022)
- [Attend and Segment: Attention Guided Active Semantic Segmentation](https://mlsanity.com/paper/2007.11548/) (2020)
- [Glimpse-Attend-and-Explore: Self-Attention for Active Visual Exploration](https://mlsanity.com/paper/2108.11717/) (2021)
- [MisConv: Convolutional Neural Networks for Missing Data](https://mlsanity.com/paper/2110.14010/) (2021)
- [Aerial View Localization with Reinforcement Learning: Towards Emulating Search-and-Rescue](https://mlsanity.com/paper/2209.03694/) (2022)
- [Visual Attention in Imaginative Agents](https://mlsanity.com/paper/2104.00177/) (2021)

## Built on by (most cited first, 2 of 2)

- [CanViT: Toward Active-Vision Foundation Models](https://mlsanity.com/paper/2603.22570/) (2026)
- [HAViT: Historical Attention Vision Transformer](https://mlsanity.com/paper/2603.18585/) (2026)

## Abstract (arXiv)

Active Visual Exploration (AVE) optimizes the utilization of robotic resources in real-world scenarios by sequentially selecting the most informative observations. However, modern methods require a high computational budget due to processing the same observations multiple times through the autoencoder transformers. As a remedy, we introduce a novel approach to AVE called TOken REcycling (TORE). It divides the encoder into extractor and aggregator components. The extractor processes each observation separately, enabling the reuse of tokens passed to the aggregator. Moreover, to further reduce the computations, we decrease the decoder to only one block. Through extensive experiments, we demonstrate that TORE outperforms state-of-the-art methods while reducing computational overhead by up to 90%.