Scaling Text-to-Image Diffusion Transformers with Representation Autoencoders

• URL: http://arxiv.org/abs/2601.16208v1
• Date: Thu, 22 Jan 2026 18:58:16 GMT
• Title: Scaling Text-to-Image Diffusion Transformers with Representation Autoencoders
• Authors: Shengbang Tong, Boyang Zheng, Ziteng Wang, Bingda Tang, Nanye Ma, Ellis Brown, Jihan Yang, Rob Fergus, Yann LeCun, Saining Xie,
• Abstract summary: 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.
• Score: 74.72147962028265
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Representation Autoencoders (RAEs) have shown distinct advantages in diffusion modeling on ImageNet by training in high-dimensional semantic latent spaces. In this work, we investigate whether this framework can scale to large-scale, freeform text-to-image (T2I) generation. We first scale RAE decoders on the frozen representation encoder (SigLIP-2) beyond ImageNet by training on web, synthetic, and text-rendering data, finding that while scale improves general fidelity, targeted data composition is essential for specific domains like text. We then rigorously stress-test the RAE design choices originally proposed for ImageNet. Our analysis reveals that scaling simplifies the framework: while dimension-dependent noise scheduling remains critical, architectural complexities such as wide diffusion heads and noise-augmented decoding offer negligible benefits at scale Building on this simplified framework, we conduct a controlled comparison of RAE against the state-of-the-art FLUX VAE across diffusion transformer scales from 0.5B to 9.8B parameters. RAEs consistently outperform VAEs during pretraining across all model scales. Further, during finetuning on high-quality datasets, VAE-based models catastrophically overfit after 64 epochs, while RAE models remain stable through 256 epochs and achieve consistently better performance. Across all experiments, RAE-based diffusion models demonstrate faster convergence and better generation quality, establishing RAEs as a simpler and stronger foundation than VAEs for large-scale T2I generation. Additionally, because both visual understanding and generation can operate in a shared representation space, the multimodal model can directly reason over generated latents, opening new possibilities for unified models.

Related papers

• One Layer Is Enough: Adapting Pretrained Visual Encoders for Image Generation [33.56782043207013]
  Feature Auto-Encoder (FAE) adapts pre-trained visual representations into low-dimensional latents suitable for generation using as little as a single attention layer.<n>FAE achieves strong performance across class-conditional and text-to-image benchmarks.
  arXiv  Detail & Related papers  (2025-12-08T18:57:26Z)
• Edit2Perceive: Image Editing Diffusion Models Are Strong Dense Perceivers [55.15722080205737]
  Edit2Perceive is a unified diffusion framework that adapts editing models for depth, normal, and matting.<n>Our single-step deterministic inference yields up to faster runtime while training on relatively small datasets.
  arXiv  Detail & Related papers  (2025-11-24T01:13:51Z)
• ScaleDiff: Higher-Resolution Image Synthesis via Efficient and Model-Agnostic Diffusion [7.233066974580282]
  Text-to-image diffusion models often exhibit degraded performance when generating images beyond their training resolution.<n>Recent training-free methods can mitigate this limitation, but they often require substantial computation or are incompatible with recent Diffusion Transformer models.<n>We propose ScaleDiff, a model-agnostic and highly efficient framework for extending the resolution of pretrained diffusion models without any additional training.
  arXiv  Detail & Related papers  (2025-10-29T17:17:32Z)
• Knowledge-Informed Neural Network for Complex-Valued SAR Image Recognition [51.03674130115878]
  We introduce the Knowledge-Informed Neural Network (KINN), a lightweight framework built upon a novel "compression-aggregation-compression" architecture.<n>KINN establishes a state-of-the-art in parameter-efficient recognition, offering exceptional generalization in data-scarce and out-of-distribution scenarios.
  arXiv  Detail & Related papers  (2025-10-23T07:12:26Z)
• Improving Progressive Generation with Decomposable Flow Matching [50.63174319509629]
  Decomposable Flow Matching (DFM) is a simple and effective framework for the progressive generation of visual media.<n>On Imagenet-1k 512px, DFM achieves 35.2% improvements in FDD scores over the base architecture and 26.4% over the best-performing baseline.
  arXiv  Detail & Related papers  (2025-06-24T17:58:02Z)
• Quantitative Comparison of Fine-Tuning Techniques for Pretrained Latent Diffusion Models in the Generation of Unseen SAR Images [0.0]
  We adapt an open-source text-to-image foundation model to the Synthetic Aperture Radar (SAR) modality.<n>We compare full fine-tuning and parameter-efficient Low-Rank Adaptation (LoRA) across the UNet diffusion backbone, the Variational Autoencoder (VAE) and the text encoders.<n>Our results show that a hybrid strategy-full UNet tuning with LoRA on the text encoders and a learned token embedding-best preserves SAR geometry and texture.
  arXiv  Detail & Related papers  (2025-06-16T09:48:01Z)
• Reconstruction vs. Generation: Taming Optimization Dilemma in Latent Diffusion Models [33.519892081718716]
  We propose aligning the latent space with pre-trained vision foundation models when training the visual tokenizers.<n>Our proposed VA-VAE significantly expands the reconstruction-generation frontier of latent diffusion models.<n>We build an enhanced DiT baseline with improved training strategies and architecture designs, termed LightningDiT.
  arXiv  Detail & Related papers  (2025-01-02T18:59:40Z)
• M-VAR: Decoupled Scale-wise Autoregressive Modeling for High-Quality Image Generation [39.97174784206976]
  We show that this scale-wise autoregressive framework can be effectively decoupled into textitintra-scale modeling We apply linear-complexity mechanisms like Mamba to substantially reduce computational overhead. Experiments demonstrate that our method outperforms existing models in both image quality and generation speed.
  arXiv  Detail & Related papers  (2024-11-15T18:54:42Z)
• DiffiT: Diffusion Vision Transformers for Image Generation [88.08529836125399]
  Vision Transformer (ViT) has demonstrated strong modeling capabilities and scalability, especially for recognition tasks. We study the effectiveness of ViTs in diffusion-based generative learning and propose a new model denoted as Diffusion Vision Transformers (DiffiT) DiffiT is surprisingly effective in generating high-fidelity images with significantly better parameter efficiency.
  arXiv  Detail & Related papers  (2023-12-04T18:57:01Z)
• Effective Invertible Arbitrary Image Rescaling [77.46732646918936]
  Invertible Neural Networks (INN) are able to increase upscaling accuracy significantly by optimizing the downscaling and upscaling cycle jointly. A simple and effective invertible arbitrary rescaling network (IARN) is proposed to achieve arbitrary image rescaling by training only one model in this work. It is shown to achieve a state-of-the-art (SOTA) performance in bidirectional arbitrary rescaling without compromising perceptual quality in LR outputs.
  arXiv  Detail & Related papers  (2022-09-26T22:22:30Z)
• Global Filter Networks for Image Classification [90.81352483076323]
  We present a conceptually simple yet computationally efficient architecture that learns long-term spatial dependencies in the frequency domain with log-linear complexity. Our results demonstrate that GFNet can be a very competitive alternative to transformer-style models and CNNs in efficiency, generalization ability and robustness.
  arXiv  Detail & Related papers  (2021-07-01T17:58:16Z)

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.