Improved Baselines with Representation Autoencoders

• URL: http://arxiv.org/abs/2605.18324v1
• Date: Mon, 18 May 2026 12:42:34 GMT
• Title: Improved Baselines with Representation Autoencoders
• Abstract summary: Representation Autoencoders (RAE) replace traditional VAE with pretrained vision encoders.<n>We find three insights which simplify and improve RAE.<n> RAEv2 leads to more than 10x faster convergence over the original RAE.
• Score: 61.47127824064028
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Representation Autoencoders (RAE) replace traditional VAE with pretrained vision encoders. In this paper, we systematically investigate several design choices and find three insights which simplify and improve RAE. First, we study a generalized formulation where the representation is defined as sum of the last k encoder layers rather than solely the final layer. This simple change greatly improves reconstruction without encoder finetuning or specialized data (e.g., text, faces). Second, we study the prevalent assumption that RAE (using pretrained representation as encoder) replaces representation alignment (REPA), which distills the same representation to intermediate layers instead. Through large-scale empirical analysis, we uncover a surprising finding: RAE and REPA exhibit complementary working mechanisms, allowing the same representation to be used as both encoder and target for intermediate diffusion layers. Finally, the original RAE struggles with classifier-free guidance (CFG) and requires training a second, weaker diffusion model for AutoGuidance (AG). We show that REPA itself can be viewed as x-prediction in RAE latent space. By simply re-parameterizing the output of the DiT model, it can provide guidance for "free". Overall, RAEv2 leads to more than 10x faster convergence over the original RAE, achieving a state-of-the-art gFID of 1.06 in just 80 epochs on ImageNet-256. On FDr^k, RAEv2 achieves a state-of-the-art 2.17 at just 80 epochs compared to the previous best 3.26 (800 epochs) without any post-training. This motivates EP_FID@k (epochs to reach unguided gFID <= k) as a measure of training efficiency. RAEv2 attains an EP_FID@2 of 35 epochs, versus 177 for the original RAE. We also validate our approach across diverse settings for text-to-image generation and navigation world models, showing consistent improvements. Code is available at https://raev2.github.io.

Related papers

• Adversarial Error Correction for Visual Autoregressive Generation [32.13829630849993]
  We propose a plug-and-play framework that enhances pre-trained Visual Autoregressive (or VAR) models.<n>Instead of a standard passive generation, AID-more introduces a proactive error-correction mechanism inspired by the adversarial feedback in GANs.<n>Results demonstrate that AID-more delivers sharper textural details and fewer structural distortions with negligible overhead.
  arXiv  Detail & Related papers  (2026-05-24T03:22:22Z)
• RAE-AR: Taming Autoregressive Models with Representation Autoencoders [61.73674018219353]
  We introduce token simplification via distribution normalization to ease modeling difficulty and improve convergence.<n>We enhance prediction by incorporating Gaussian noise injection during training to mitigate exposure bias.<n>This work paves the way for a more unified architecture across visual understanding and generative modeling.
  arXiv  Detail & Related papers  (2026-04-02T02:39:28Z)
• DINO-SAE: DINO Spherical Autoencoder for High-Fidelity Image Reconstruction and Generation [47.409626500688866]
  We present the DINO Spherical Autoencoder (DINO-SAE), a framework that bridges semantic representation and pixel-level reconstruction.<n>Our approach achieves state-of-the-art reconstruction quality, reaching 0.37 rFID and 26.2 dB PSNR, while maintaining strong semantic alignment to the pretrained VFM.
  arXiv  Detail & Related papers  (2026-01-30T12:25:34Z)
• Scaling Text-to-Image Diffusion Transformers with Representation Autoencoders [74.72147962028265]
  Representation Autoencoders (RAEs) have shown distinct advantages in diffusion modeling on ImageNet.<n>We investigate whether this framework can scale to large-scale, freeform text-to-image (T2I) generation.
  arXiv  Detail & Related papers  (2026-01-22T18:58:16Z)
• What matters for Representation Alignment: Global Information or Spatial Structure? [64.67092609921816]
  Representation alignment (REPA) guides generative training by distilling representations from a strong, pretrained vision encoder to intermediate diffusion features.<n>We investigate a fundamental question: what aspect of the target representation matters for generation, its textitglobal revisionsemantic information.<n>We replace the standard projection layer in REPA with a simple convolution layer and introduce a spatial normalization layer for the external representation.
  arXiv  Detail & Related papers  (2025-12-11T16:39:53Z)
• REAR: Rethinking Visual Autoregressive Models via Generator-Tokenizer Consistency Regularization [130.46612643194973]
  reAR is a simple training strategy introducing a token-wise regularization objective.<n>On ImageNet, it reduces gFID from 3.02 to 1.86 and improves IS to 316.9 using a standardization-based tokenizer.<n>When applied to advanced tokenizers, it achieves a gFID of 1.42 with only 177M parameters, matching the performance with larger state-of-the-art diffusion models (675M)
  arXiv  Detail & Related papers  (2025-10-06T02:48:13Z)
• NVAE: A Deep Hierarchical Variational Autoencoder [102.29977384039805]
  We propose a deep hierarchical VAE built for image generation using depth-wise separable convolutions and batch normalization. We show that NVAE achieves state-of-the-art results among non-autoregressive likelihood-based models. To the best of our knowledge, NVAE is the first successful VAE applied to natural images as large as 256$times $256 pixels.
  arXiv  Detail & Related papers  (2020-07-08T04:56:56Z)

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.