TextLDM: Language Modeling with Continuous Latent Diffusion

• URL: http://arxiv.org/abs/2605.07748v1
• Date: Fri, 08 May 2026 13:54:34 GMT
• Title: TextLDM: Language Modeling with Continuous Latent Diffusion
• Abstract summary: Diffusion Transformers (DiT) trained with flow matching in a VAE latent space have unified visual generation across images and videos.<n>We propose TextLDM, which transfers the visual latent diffusion recipe to text generation with minimal architectural modification.
• Score: 89.69255520673248
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Diffusion Transformers (DiT) trained with flow matching in a VAE latent space have unified visual generation across images and videos. A natural next step toward a single architecture for both generation (visual synthesis) and understanding (text generation) is to apply this framework to language modeling. We propose TextLDM, which transfers the visual latent diffusion recipe to text generation with minimal architectural modification. A Transformer-based VAE maps discrete tokens to continuous latents, enhanced by Representation Alignment (REPA) with a frozen pretrained language model to produce representations effective for conditional denoising. A standard DiT then performs flow matching in this latent space, identical in architecture to its visual counterpart. The central challenge we address is obtaining high-quality continuous text representations: we find that reconstruction fidelity alone is insufficient, and that aligning latent features with a pretrained language model via REPA is critical for downstream generation quality. Trained from scratch on OpenWebText2, TextLDM substantially outperforms prior diffusion language models and matches GPT-2 under the same settings. Our results establish that the visual DiT recipe transfers effectively to language, taking a concrete step toward unified diffusion architectures for multimodal generation and understanding.

Related papers

• AURORA-LM: Autoencoding Unified Representation for Continuous-Latent Diffusion Language Modeling [60.04959047453115]
  AURORA-LM is a continuous-latent language model that separates the construction of a decodable text representation from the modeling of its distribution.<n>AURORA-LM achieves the strongest performance among evaluated continuous and diffusion-based language models on OpenWebText free generation and XSum summarization.
  arXiv  Detail & Related papers  (2026-08-03T17:59:50Z)
• How to Train Your Latent Diffusion Language Model Jointly With the Latent Space [76.9057986588963]
  Latent diffusion models offer an attractive alternative to discrete diffusion for non-autoregressive text generation.<n>We present the Latent Diffusion Language Model (LDLM), in which the latent encoder, diffusion model, and decoder are trained jointly.<n>On OpenWebText and LM1B, LDLM achieves better generation performance than existing discrete and continuous diffusion language models.
  arXiv  Detail & Related papers  (2026-05-08T16:05:19Z)
• Continuous Latent Diffusion Language Model [48.974403879186916]
  Large language models have achieved remarkable success under the autoregressive paradigm.<n>Existing alternatives still struggle to jointly achieve generation efficiency, scalable representation learning, and effective global semantic modeling.<n>We propose Cola DLM, a hierarchical latent diffusion language model that frames text generation through hierarchical information decomposition.
  arXiv  Detail & Related papers  (2026-05-07T16:44:56Z)
• LaDiC: Are Diffusion Models Really Inferior to Autoregressive Counterparts for Image-to-Text Generation? [10.72249123249003]
  We revisit diffusion models, highlighting their capacity for holistic context modeling and parallel decoding. We introduce a novel architecture, LaDiC, which utilizes a split BERT to create a dedicated latent space for captions. LaDiC achieves state-of-the-art performance for diffusion-based methods on the MS dataset with 38.2 BLEU@4 and 126.2 CIDEr.
  arXiv  Detail & Related papers  (2024-04-16T17:47:16Z)
• Learning from Mistakes: Iterative Prompt Relabeling for Text-to-Image Diffusion Model Training [33.51524424536508]
  Iterative Prompt Relabeling (IPR) is a novel algorithm that aligns images to text through iterative image sampling and prompt relabeling with feedback.<n>We conduct thorough experiments on SDv2 and SDXL, testing their capability to follow instructions on spatial relations.
  arXiv  Detail & Related papers  (2023-12-23T11:10:43Z)
• Reason out Your Layout: Evoking the Layout Master from Large Language Models for Text-to-Image Synthesis [47.27044390204868]
  We introduce a novel approach to improving T2I diffusion models using Large Language Models (LLMs) as layout generators. Our experiments demonstrate significant improvements in image quality and layout accuracy.
  arXiv  Detail & Related papers  (2023-11-28T14:51:13Z)
• LLM-grounded Diffusion: Enhancing Prompt Understanding of Text-to-Image Diffusion Models with Large Language Models [62.75006608940132]
  This work proposes to enhance prompt understanding capabilities in text-to-image diffusion models. Our method leverages a pretrained large language model for grounded generation in a novel two-stage process. Our method significantly outperforms the base diffusion model and several strong baselines in accurately generating images.
  arXiv  Detail & Related papers  (2023-05-23T03:59:06Z)
• Semantic-Conditional Diffusion Networks for Image Captioning [116.86677915812508]
  We propose a new diffusion model based paradigm tailored for image captioning, namely Semantic-Conditional Diffusion Networks (SCD-Net) In SCD-Net, multiple Diffusion Transformer structures are stacked to progressively strengthen the output sentence with better visional-language alignment and linguistical coherence. Experiments on COCO dataset demonstrate the promising potential of using diffusion models in the challenging image captioning task.
  arXiv  Detail & Related papers  (2022-12-06T16:08:16Z)
• eDiffi: Text-to-Image Diffusion Models with an Ensemble of Expert Denoisers [87.52504764677226]
  Large-scale diffusion-based generative models have led to breakthroughs in text-conditioned high-resolution image synthesis. We train an ensemble of text-to-image diffusion models specialized for different stages synthesis. Our ensemble of diffusion models, called eDiffi, results in improved text alignment while maintaining the same inference cost.
  arXiv  Detail & Related papers  (2022-11-02T17:43:04Z)

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.