Simplifying DINO via Coding Rate Regularization

• URL: http://arxiv.org/abs/2502.10385v1
• Date: Fri, 14 Feb 2025 18:58:04 GMT
• Title: Simplifying DINO via Coding Rate Regularization
• Abstract summary: DINO and DINOv2 are two model families being widely used to learn representations from unlabeled imagery data at large scales.<n>This work highlights the potential of using simplifying design principles to improve the empirical practice of deep learning.
• Score: 74.88963795406733
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: DINO and DINOv2 are two model families being widely used to learn representations from unlabeled imagery data at large scales. Their learned representations often enable state-of-the-art performance for downstream tasks, such as image classification and segmentation. However, they employ many empirically motivated design choices and their training pipelines are highly complex and unstable -- many hyperparameters need to be carefully tuned to ensure that the representations do not collapse -- which poses considerable difficulty to improving them or adapting them to new domains. In this work, we posit that we can remove most such-motivated idiosyncrasies in the pre-training pipelines, and only need to add an explicit coding rate term in the loss function to avoid collapse of the representations. As a result, we obtain highly simplified variants of the DINO and DINOv2 which we call SimDINO and SimDINOv2, respectively. Remarkably, these simplified models are more robust to different design choices, such as network architecture and hyperparameters, and they learn even higher-quality representations, measured by performance on downstream tasks, offering a Pareto improvement over the corresponding DINO and DINOv2 models. This work highlights the potential of using simplifying design principles to improve the empirical practice of deep learning.

Related papers

• Where Does Generative Difficulty Reside? An Empirical Study of Target Representations [62.54876287800644]
  The target representation defines the distribution an image generator must learn, yet it is often treated as an interchangeable interface.<n>We study raw pixels, SD-VAE latents and DINOv2 as well as MAE representation-autoencoder features within a unified masked autoregressive rectified-flow model.<n>Our results show that compression, reconstruction fidelity, token dimensionality, and visible semantic clustering do not individually predict generative behavior.
  arXiv  Detail & Related papers  (2026-08-01T12:32:31Z)
• MEPA: Multi-Scale Representation Alignment for Visual Autoregressive Modeling with Mixture of Experts [41.67455261906954]
  We introduce a scale-aware token-routed Mixture of Experts (MoE) architecture, allowing scale-adaptive expert selection.<n>Unlike naive alignment, we analyse and design a residual feature aggregation scheme tailored to the Visual AutoRegressive modeling paradigm.<n>Our model achieves a superior FID compared to the dense baseline on the ImageNet 256*256 benchmark.
  arXiv  Detail & Related papers  (2026-07-01T03:11:21Z)
• Interpretability Without Tradeoffs: Disentangling Polysemanticity At Equal Predictive Performance [58.49623467782041]
  ELUDe is a method for improving the interpretability of deep neural networks.<n>It breaks latent representations into clear, inspectable sub-units that behave like interpretable features.<n>It requires no explicit training, no labels, and can be applied to pretrained models.
  arXiv  Detail & Related papers  (2026-05-29T13:39:09Z)
• KVNN: Learnable Multi-Kernel Volterra Neural Networks [6.550684951976901]
  kernelized Volterra Neural Network (kVNN)<n>Experiments on two representative tasks: video action recognition and image denoising.<n>Results are maintained even when trained from scratch without large-scale pretraining.
  arXiv  Detail & Related papers  (2026-04-16T15:18:48Z)
• Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think [72.48325960659822]
  One main bottleneck in training large-scale diffusion models for generation lies in effectively learning these representations.<n>We study this by introducing a straightforward regularization called REPresentation Alignment (REPA), which aligns the projections of noisy input hidden states in denoising networks with clean image representations obtained from external, pretrained visual encoders.<n>The results are striking: our simple strategy yields significant improvements in both training efficiency and generation quality when applied to popular diffusion and flow-based transformers, such as DiTs and SiTs.
  arXiv  Detail & Related papers  (2024-10-09T14:34:53Z)
• Code Representation Learning At Scale [75.04686476303436]
  We fuel code representation learning with a vast amount of code data via a two-stage pretraining scheme. We first train the encoders via a mix that leverages both randomness in masking language modeling and the structure aspect of programming language. We then enhance the representations via contrastive learning with hard negative and hard positive constructed in an unsupervised manner.
  arXiv  Detail & Related papers  (2024-02-02T22:19:15Z)
• Improving Long-Horizon Imitation Through Instruction Prediction [93.47416552953075]
  In this work, we explore the use of an often unused source of auxiliary supervision: language. Inspired by recent advances in transformer-based models, we train agents with an instruction prediction loss that encourages learning temporally extended representations that operate at a high level of abstraction. In further analysis we find that instruction modeling is most important for tasks that require complex reasoning, while understandably offering smaller gains in environments that require simple plans.
  arXiv  Detail & Related papers  (2023-06-21T20:47:23Z)
• Improving the Robustness of Neural Multiplication Units with Reversible Stochasticity [2.4278445972594525]
  Multilayer Perceptrons struggle to learn certain simple arithmetic tasks. Specialist neural NMU (sNMU) is proposed to apply reversibleity, encouraging avoidance of such optima.
  arXiv  Detail & Related papers  (2022-11-10T14:56:37Z)
• Randomized Histogram Matching: A Simple Augmentation for Unsupervised Domain Adaptation in Overhead Imagery [3.187381965457262]
  We propose a fast real-time unsupervised training augmentation technique, termed randomized histogram matching (RHM) RHM consistently yields similar or superior performance compared to state-of-the-art unsupervised domain adaptation approaches. RHM also offers substantially better performance than other comparably simple approaches that are widely used for overhead imagery.
  arXiv  Detail & Related papers  (2021-04-28T21:59:54Z)
• Dissecting Supervised Constrastive Learning [24.984074794337157]
  Minimizing cross-entropy over the softmax scores of a linear map composed with a high-capacity encoder is arguably the most popular choice for training neural networks on supervised learning tasks. We show that one can directly optimize the encoder instead, to obtain equally (or even more) discriminative representations via a supervised variant of a contrastive objective.
  arXiv  Detail & Related papers  (2021-02-17T15:22:38Z)
• Style Normalization and Restitution for DomainGeneralization and Adaptation [88.86865069583149]
  An effective domain generalizable model is expected to learn feature representations that are both generalizable and discriminative. In this paper, we design a novel Style Normalization and Restitution module (SNR) to ensure both high generalization and discrimination capability of the networks.
  arXiv  Detail & Related papers  (2021-01-03T09:01:39Z)
• Recurrent Multi-view Alignment Network for Unsupervised Surface Registration [79.72086524370819]
  Learning non-rigid registration in an end-to-end manner is challenging due to the inherent high degrees of freedom and the lack of labeled training data. We propose to represent the non-rigid transformation with a point-wise combination of several rigid transformations. We also introduce a differentiable loss function that measures the 3D shape similarity on the projected multi-view 2D depth images.
  arXiv  Detail & Related papers  (2020-11-24T14:22:42Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.