# Leveraging Registers in Vision Transformers for Robust Adaptation

arXiv: https://arxiv.org/abs/2501.04784 · HTML: https://mlsanity.com/paper/2501.04784/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 25 references that are in the map.
- First posted on arXiv 8 Jan 2025.

## Builds on (most cited first, 20 of 20)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks](https://mlsanity.com/paper/1610.02136/) (2016)
- [Natural Adversarial Examples](https://mlsanity.com/paper/1907.07174/) (2019)
- [Energy-based Out-of-distribution Detection](https://mlsanity.com/paper/2010.03759/) (2020)
- [Learning Robust Global Representations by Penalizing Local Predictive Power](https://mlsanity.com/paper/1905.13549/) (2019)
- [Scaling Vision Transformers](https://mlsanity.com/paper/2106.04560/) (2021)
- [Vision Transformers Need Registers](https://mlsanity.com/paper/2309.16588/) (2023)
- [DeiT III: Revenge of the ViT](https://mlsanity.com/paper/2204.07118/) (2022)
- [Massive Activations in Large Language Models](https://mlsanity.com/paper/2402.17762/) (2024)
- [All Bark and No Bite: Rogue Dimensions in Transformer Language Models Obscure Representational Quality](https://mlsanity.com/paper/2109.04404/) (2021)
- [BERT Busters: Outlier Dimensions that Disrupt Transformers](https://mlsanity.com/paper/2105.06990/) (2021)
- [An Inverse Scaling Law for CLIP Training](https://mlsanity.com/paper/2305.07017/) (2023)
- [Mamba-R: Vision Mamba ALSO Needs Registers](https://mlsanity.com/paper/2405.14858/) (2024)
- [Learning useful representations for shifting tasks and distributions](https://mlsanity.com/paper/2212.07346/) (2022)
- [Unveiling A Core Linguistic Region in Large Language Models](https://mlsanity.com/paper/2310.14928/) (2023)
- [Understanding and Minimising Outlier Features in Neural Network Training](https://mlsanity.com/paper/2405.19279/) (2024)

## Built on by (most cited first, 3 of 3)

- [Understanding vision transformer robustness through the lens of out-of-distribution detection](https://mlsanity.com/paper/2602.01459/) (2026)
- [Zero-Ablation Overstates Register Content Dependence in DINO Vision Transformers](https://mlsanity.com/paper/2604.14433/) (2026)
- [RegisterBridgeMM: A Register-Centric Framework for RGB-Infrared Object Detection](https://mlsanity.com/paper/2608.04833/) (2026)

## Abstract (arXiv)

Vision Transformers (ViTs) have shown success across a variety of tasks due to their ability to capture global image representations. Recent studies have identified the existence of high-norm tokens in ViTs, which can interfere with unsupervised object discovery. To address this, the use of "registers" which are additional tokens that isolate high norm patch tokens while capturing global image-level information has been proposed. While registers have been studied extensively for object discovery, their generalization properties particularly in out-of-distribution (OOD) scenarios, remains underexplored. In this paper, we examine the utility of register token embeddings in providing additional features for improving generalization and anomaly rejection. To that end, we propose a simple method that combines the special CLS token embedding commonly employed in ViTs with the average-pooled register embeddings to create feature representations which are subsequently used for training a downstream classifier. We find that this enhances OOD generalization and anomaly rejection, while maintaining in-distribution (ID) performance. Extensive experiments across multiple ViT backbones trained with and without registers reveal consistent improvements of 2-4% in top-1 OOD accuracy and a 2-3% reduction in false positive rates for anomaly detection. Importantly, these gains are achieved without additional computational overhead.