Instella-T2I: Pushing the Limits of 1D Discrete Latent Space Image Generation

• URL: http://arxiv.org/abs/2506.21022v1
• Date: Thu, 26 Jun 2025 05:48:36 GMT
• Title: Instella-T2I: Pushing the Limits of 1D Discrete Latent Space Image Generation
• Authors: Ze Wang, Hao Chen, Benran Hu, Jiang Liu, Ximeng Sun, Jialian Wu, Yusheng Su, Xiaodong Yu, Emad Barsoum, Zicheng Liu,
• Abstract summary: We introduce 1D binary image latents for compact discrete representation of images.<n>Our approach preserves high-resolution details while maintaining the compactness of 1D latents.<n>Our text-to-image models are the first to achieve competitive performance in both diffusion and auto-regressive generation.
• Score: 27.795313102716726
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Image tokenization plays a critical role in reducing the computational demands of modeling high-resolution images, significantly improving the efficiency of image and multimodal understanding and generation. Recent advances in 1D latent spaces have reduced the number of tokens required by eliminating the need for a 2D grid structure. In this paper, we further advance compact discrete image representation by introducing 1D binary image latents. By representing each image as a sequence of binary vectors, rather than using traditional one-hot codebook tokens, our approach preserves high-resolution details while maintaining the compactness of 1D latents. To the best of our knowledge, our text-to-image models are the first to achieve competitive performance in both diffusion and auto-regressive generation using just 128 discrete tokens for images up to 1024x1024, demonstrating up to a 32-fold reduction in token numbers compared to standard VQ-VAEs. The proposed 1D binary latent space, coupled with simple model architectures, achieves marked improvements in speed training and inference speed. Our text-to-image models allow for a global batch size of 4096 on a single GPU node with 8 AMD MI300X GPUs, and the training can be completed within 200 GPU days. Our models achieve competitive performance compared to modern image generation models without any in-house private training data or post-training refinements, offering a scalable and efficient alternative to conventional tokenization methods.

Related papers

• Token-Shuffle: Towards High-Resolution Image Generation with Autoregressive Models [92.18057318458528]
  Token-Shuffle is a novel method that reduces the number of image tokens in Transformer.<n>Our strategy requires no additional pretrained text-encoder and enables MLLMs to support extremely high-resolution image synthesis.<n>In GenAI-benchmark, our 2.7B model achieves 0.77 overall score on hard prompts, outperforming AR models LlamaGen by 0.18 and diffusion models LDM by 0.15.
  arXiv  Detail & Related papers  (2025-04-24T17:59:56Z)
• When Worse is Better: Navigating the compression-generation tradeoff in visual tokenization [92.17160980120404]
  We introduce Causally Regularized Tokenization (CRT), which uses knowledge of the stage 2 generation modeling procedure to embed useful inductive biases in stage 1 latents.<n>CRT makes stage 1 reconstruction performance worse, but makes stage 2 generation performance better by making the tokens easier to model.<n>We match state-of-the-art discrete autoregressive ImageNet generation (2.18 FID) with less than half the tokens per image.
  arXiv  Detail & Related papers  (2024-12-20T20:32:02Z)
• SoftVQ-VAE: Efficient 1-Dimensional Continuous Tokenizer [45.720721058671856]
  SoftVQ-VAE is a continuous image tokenizer that leverages soft categorical posteriors to aggregate multiple codewords into each latent token.<n>Our approach compresses 256x256 and 512x512 images using as few as 32 or 64 1-dimensional tokens.<n>Remarkably, SoftVQ-VAE improves inference throughput by up to 18x for generating 256x256 images and 55x for 512x512 images.
  arXiv  Detail & Related papers  (2024-12-14T20:29:29Z)
• MaskBit: Embedding-free Image Generation via Bit Tokens [54.827480008982185]
  We present an empirical and systematic examination of VQGANs, leading to a modernized VQGAN.<n>Second contribution demonstrates that embedding-free image generation using bit tokens achieves a new state-of-the-art FID of 1.52 on the ImageNet 256x256 benchmark.
  arXiv  Detail & Related papers  (2024-09-24T16:12:12Z)
• An Image is Worth 32 Tokens for Reconstruction and Generation [54.24414696392026]
  Transformer-based 1-Dimensional Tokenizer (TiTok) is an innovative approach that tokenizes images into 1D latent sequences. TiTok achieves competitive performance to state-of-the-art approaches. Our best-performing variant can significantly surpasses DiT-XL/2 (gFID 2.13 vs. 3.04) while still generating high-quality samples 74x faster.
  arXiv  Detail & Related papers  (2024-06-11T17:59:56Z)
• SDXS: Real-Time One-Step Latent Diffusion Models with Image Conditions [5.100085108873068]
  We present two models, SDXS-512 and SDXS-1024, achieving inference speeds of approximately 100 FPS (30x faster than SD v1.5) and 30 FPS (60x faster than SDXL) on a single GPU. Our training approach offers promising applications in image-conditioned control, facilitating efficient image-to-image translation.
  arXiv  Detail & Related papers  (2024-03-25T11:16:23Z)
• Emage: Non-Autoregressive Text-to-Image Generation [63.347052548210236]
  Non-autoregressive text-to-image models efficiently generate hundreds of image tokens in parallel. Our model with 346M parameters generates an image of 256$times$256 with about one second on one V100 GPU.
  arXiv  Detail & Related papers  (2023-12-22T10:01:54Z)
• Binary Latent Diffusion [36.70550531181131]
  We show that a binary latent space can be explored for compact yet expressive image representations. We present both conditional and unconditional image generation experiments with multiple datasets. The proposed framework can be seamlessly scaled to $1024 times 1024$ high-resolution image generation without resorting to latent hierarchy or multi-stage refinements.
  arXiv  Detail & Related papers  (2023-04-10T19:03:28Z)
• Locally Masked Convolution for Autoregressive Models [107.4635841204146]
  LMConv is a simple modification to the standard 2D convolution that allows arbitrary masks to be applied to the weights at each location in the image. We learn an ensemble of distribution estimators that share parameters but differ in generation order, achieving improved performance on whole-image density estimation.
  arXiv  Detail & Related papers  (2020-06-22T17:59:07Z)

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.