# The Principle of Diversity: Training Stronger Vision Transformers Calls for Reducing All Levels of Redundancy

arXiv: https://arxiv.org/abs/2203.06345 · HTML: https://mlsanity.com/paper/2203.06345/

## Key facts

- Cited by 35 papers in the map (data through 14 Oct 2026).
- Builds on 68 of its 78 references that are in the map.
- First posted on arXiv 12 Mar 2022.

## Builds on (most cited first, 30 of 68)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [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)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [End-to-End Object Detection with Transformers](https://mlsanity.com/paper/2005.12872/) (2020)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://mlsanity.com/paper/2010.04159/) (2020)
- [FaceNet: A Unified Embedding for Face Recognition and Clustering](https://mlsanity.com/paper/1503.03832/) (2015)
- [Supervised Contrastive Learning](https://mlsanity.com/paper/2004.11362/) (2020)
- [ViLBERT: Pretraining Task-Agnostic Visiolinguistic Representations for Vision-and-Language Tasks](https://mlsanity.com/paper/1908.02265/) (2019)
- [Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions](https://mlsanity.com/paper/2102.12122/) (2021)
- [ViViT: A Video Vision Transformer](https://mlsanity.com/paper/2103.15691/) (2021)
- [LXMERT: Learning Cross-Modality Encoder Representations from Transformers](https://mlsanity.com/paper/1908.07490/) (2019)
- [Point Transformer](https://mlsanity.com/paper/2011.00931/) (2020)
- [Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hypersphere](https://mlsanity.com/paper/2005.10242/) (2020)
- [VisualBERT: A Simple and Performant Baseline for Vision and Language](https://mlsanity.com/paper/1908.03557/) (2019)
- [Unified Perceptual Parsing for Scene Understanding](https://mlsanity.com/paper/1807.10221/) (2018)
- [Oscar: Object-Semantics Aligned Pre-training for Vision-Language Tasks](https://mlsanity.com/paper/2004.06165/) (2020)
- [Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNet](https://mlsanity.com/paper/2101.11986/) (2021)
- [VL-BERT: Pre-training of Generic Visual-Linguistic Representations](https://mlsanity.com/paper/1908.08530/) (2019)
- [CvT: Introducing Convolutions to Vision Transformers](https://mlsanity.com/paper/2103.15808/) (2021)
- [Pre-Trained Image Processing Transformer](https://mlsanity.com/paper/2012.00364/) (2020)
- [CrossViT: Cross-Attention Multi-Scale Vision Transformer for Image Classification](https://mlsanity.com/paper/2103.14899/) (2021)
- [Image Transformer](https://mlsanity.com/paper/1802.05751/) (2018)
- [Going deeper with Image Transformers](https://mlsanity.com/paper/2103.17239/) (2021)
- [Transformer in Transformer](https://mlsanity.com/paper/2103.00112/) (2021)

## Built on by (most cited first, 30 of 35)

- [Token Contrast for Weakly-Supervised Semantic Segmentation](https://mlsanity.com/paper/2303.01267/) (2023)
- [Sparse MoE as the New Dropout: Scaling Dense and Self-Slimmable Transformers](https://mlsanity.com/paper/2303.01610/) (2023)
- [SHViT: Single-Head Vision Transformer with Memory Efficient Macro Design](https://mlsanity.com/paper/2401.16456/) (2024)
- [Visual Tuning](https://mlsanity.com/paper/2305.06061/) (2023)
- [Measure-to-measure interpolation using Transformers](https://mlsanity.com/paper/2411.04551/) (2024)
- [PPT: token-Pruned Pose Transformer for monocular and multi-view human pose estimation](https://mlsanity.com/paper/2209.08194/) (2022)
- [How do Transformers perform In-Context Autoregressive Learning?](https://mlsanity.com/paper/2402.05787/) (2024)
- [Expedited Training of Visual Conditioned Language Generation via Redundancy Reduction](https://mlsanity.com/paper/2310.03291/) (2023)
- [Generalizing and Decoupling Neural Collapse via Hyperspherical Uniformity Gap](https://mlsanity.com/paper/2303.06484/) (2023)
- [Task Arithmetic in Trust Region: A Training-Free Model Merging Approach to Navigate Knowledge Conflicts](https://mlsanity.com/paper/2501.15065/) (2025)
- [Graph Convolutions Enrich the Self-Attention in Transformers!](https://mlsanity.com/paper/2312.04234/) (2023)
- [The Lighter The Better: Rethinking Transformers in Medical Image Segmentation Through Adaptive Pruning](https://mlsanity.com/paper/2206.14413/) (2022)
- [Reparameterized LLM Training via Orthogonal Equivalence Transformation](https://mlsanity.com/paper/2506.08001/) (2025)
- [Data Level Lottery Ticket Hypothesis for Vision Transformers](https://mlsanity.com/paper/2211.01484/) (2022)
- [Focus on the Core: Efficient Attention via Pruned Token Compression for Document Classification](https://mlsanity.com/paper/2406.01283/) (2024)
- [Sparse Imagination for Efficient Visual World Model Planning](https://mlsanity.com/paper/2506.01392/) (2025)
- [Data-independent Module-aware Pruning for Hierarchical Vision Transformers](https://mlsanity.com/paper/2404.13648/) (2024)
- [Token-Label Alignment for Vision Transformers](https://mlsanity.com/paper/2210.06455/) (2022)
- [Aligning Non-Causal Factors for Transformer-Based Source-Free Domain Adaptation](https://mlsanity.com/paper/2311.16294/) (2023)
- [Calibrating Undisciplined Over-Smoothing in Transformer for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/2305.03112/) (2023)
- [Attacking Transformers with Feature Diversity Adversarial Perturbation](https://mlsanity.com/paper/2403.07942/) (2024)
- [Merging Smarter, Generalizing Better: Enhancing Model Merging on OOD Data](https://mlsanity.com/paper/2506.09093/) (2025)
- [Frequency-Aware Token Reduction for Efficient Vision Transformer](https://mlsanity.com/paper/2511.21477/) (2025)
- [Merging of neural networks](https://mlsanity.com/paper/2204.09973/) (2022)
- [Hyper-SET: Designing Transformers via Hyperspherical Energy Minimization](https://mlsanity.com/paper/2502.11646/) (2025)
- [Kolmogorov-Arnold Attention: Is Learnable Attention Better For Vision Transformers?](https://mlsanity.com/paper/2503.10632/) (2025)
- [Image Recognition with Online Lightweight Vision Transformer: A Survey](https://mlsanity.com/paper/2505.03113/) (2025)
- [ConceptHash: Interpretable Fine-Grained Hashing via Concept Discovery](https://mlsanity.com/paper/2406.08457/) (2024)
- [Down-Sampling Inter-Layer Adapter for Parameter and Computation Efficient Ultra-Fine-Grained Image Recognition](https://mlsanity.com/paper/2409.11051/) (2024)
- [Recent Advances of Local Mechanisms in Computer Vision: A Survey and Outlook of Recent Work](https://mlsanity.com/paper/2306.01929/) (2023)

## Abstract (arXiv)

Vision transformers (ViTs) have gained increasing popularity as they are commonly believed to own higher modeling capacity and representation flexibility, than traditional convolutional networks. However, it is questionable whether such potential has been fully unleashed in practice, as the learned ViTs often suffer from over-smoothening, yielding likely redundant models. Recent works made preliminary attempts to identify and alleviate such redundancy, e.g., via regularizing embedding similarity or re-injecting convolution-like structures. However, a "head-to-toe assessment" regarding the extent of redundancy in ViTs, and how much we could gain by thoroughly mitigating such, has been absent for this field. This paper, for the first time, systematically studies the ubiquitous existence of redundancy at all three levels: patch embedding, attention map, and weight space. In view of them, we advocate a principle of diversity for training ViTs, by presenting corresponding regularizers that encourage the representation diversity and coverage at each of those levels, that enabling capturing more discriminative information. Extensive experiments on ImageNet with a number of ViT backbones validate the effectiveness of our proposals, largely eliminating the observed ViT redundancy and significantly boosting the model generalization. For example, our diversified DeiT obtains 0.70%~1.76% accuracy boosts on ImageNet with highly reduced similarity. Our codes are fully available in https://github.com/VITA-Group/Diverse-ViT.