mlsanityData to 7 Oct 2026

ELF: Embedded Language Flows

arXiv 2605.10938 ↗First posted 11 May 2026cs.CL cs.AI cs.LGcited by 27

On the map: Diffusion models → Discrete and text diffusion

Abstract (arXiv)

Diffusion and flow-based models have become the de facto approaches for generating continuous data, e.g., in domains such as images and videos. Their success has attracted growing interest in applying them to language modeling. Unlike their image-domain counterparts, today's leading diffusion language models (DLMs) primarily operate over discrete tokens. In this paper, we show that continuous DLMs can be made effective with minimal adaptation to the discrete domain. We propose Embedded Language Flows (ELF), a class of diffusion models in continuous embedding space based on continuous-time Flow Matching. Unlike existing DLMs, ELF predominantly stays within the continuous embedding space until the final time step, where it maps to discrete tokens using a shared-weight network. This formulation makes it straightforward to adapt established techniques from image-domain diffusion models, e.g., classifier-free guidance (CFG). Experiments show that ELF substantially outperforms leading discrete and continuous DLMs, achieving better generation quality with fewer sampling steps. These results suggest that ELF offers a promising path toward effective continuous DLMs.

Key facts. Cited by 27 papers in the map (data through 7 Oct 2026). Placed in Discrete and text diffusion, one of 14 directions in Diffusion models. Builds on 81 of its 93 references that are in the map. First posted on arXiv 11 May 2026.

Builds on

81 of 93 references.

Closest ancestors · 43

Same direction, newest first

  1. 2026How to Train Your Latent Diffusion Language Model Jointly With the Latent Space8 May
  2. 2026Continuous Latent Diffusion Language Model7 May
  3. 2026LangFlow: Continuous Diffusion Rivals Discrete in Language Modeling13 Apr
  4. 2026CoDAR: Continuous Diffusion Language Models are More Powerful Than You Think3 Mar
  5. 2026The Diffusion Duality, Chapter II: Psi-Samplers24 Feb
  6. 2026Flow Map Language Models: One-step Language Modeling via Continuous Denoising18 Feb
  7. 2026Scaling Beyond Masked Diffusion Language Models16 Feb
  8. 2025Encoder-Decoder Diffusion Language Models for Efficient Training and Inference26 Oct
  9. 2025CANDI: Hybrid Discrete-Continuous Diffusion Models26 Oct
  10. 2025Coevolutionary Continuous Discrete Diffusion: Make Your Diffusion Language Model a Latent Reasoner3 Oct
  11. 2025Continuously Augmented Discrete Diffusion model for Categorical Generative Modeling1 Oct
  12. 2025Dream 7B: Diffusion Large Language Models21 Aug
31 more
  1. 2025A Survey on Diffusion Language Models14 Aug
  2. 2025Seed Diffusion: A Large-Scale Diffusion Language Model with High-Speed Inference4 Aug
  3. 2025Cosmos: Compressed and Smooth Latent Space for Text Diffusion Modeling26 Jun
  4. 2025DiffuCoder: Understanding and Improving Masked Diffusion Models for Code Generation25 Jun
  5. 2025The Diffusion Duality12 Jun
  6. 2025Remasking Discrete Diffusion Models with Inference-Time Scaling1 Mar
  7. 2025TESS 2: A Large-Scale Generalist Diffusion Language Model19 Feb
  8. 2025Continuous Diffusion Model for Language Modeling17 Feb
  9. 2025Large Language Diffusion Models14 Feb
  10. 2024Beyond Autoregression: Fast LLMs via Self-Distillation Through Time28 Oct
  11. 2024Diffusion Guided Language Modeling8 Aug
  12. 2024Simple and Effective Masked Diffusion Language Models11 Jun
  13. 2024Simplified and Generalized Masked Diffusion for Discrete Data6 Jun
  14. 2024TEncDM: Understanding the Properties of the Diffusion Model in the Space of Language Model Encodings29 Feb
  15. 2023Discrete Diffusion Modeling by Estimating the Ratios of the Data Distribution25 Oct
  16. 2023InfoDiffusion: Information Entropy Aware Diffusion Process for Non-Autoregressive Text Generation18 Oct
  17. 2023DiffuSeq-v2: Bridging Discrete and Continuous Text Spaces for Accelerated Seq2Seq Diffusion Models9 Oct
  18. 2023PLANNER: Generating Diversified Paragraph via Latent Language Diffusion Model5 Jun
  19. 2023Likelihood-Based Diffusion Language Models30 May
  20. 2023AR-Diffusion: Auto-Regressive Diffusion Model for Text Generation16 May
  21. 2023TESS: Text-to-Text Self-Conditioned Simplex Diffusion15 May
  22. 2023DINOISER: Diffused Conditional Sequence Learning by Manipulating Noises20 Feb
  23. 2022Text Generation with Diffusion Language Models: A Pre-training Approach with Continuous Paragraph Denoise22 Dec
  24. 2022SeqDiffuSeq: Text Diffusion with Encoder-Decoder Transformers20 Dec
  25. 2022Latent Diffusion for Language Generation19 Dec
  26. 2022Empowering Diffusion Models on the Embedding Space for Text Generation19 Dec
  27. 2022Continuous diffusion for categorical data28 Nov
  28. 2022DiffusionBERT: Improving Generative Masked Language Models with Diffusion Models28 Nov
  29. 2022Self-conditioned Embedding Diffusion for Text Generation8 Nov
  30. 2022SSD-LM: Semi-autoregressive Simplex-based Diffusion Language Model for Text Generation and Modular Control31 Oct
  31. 2022DiffuSeq: Sequence to Sequence Text Generation with Diffusion Models17 Oct

Other references · 38

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  3. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  4. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  5. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  6. 2022Scalable Diffusion Models with Transformerscited by 5,931
  7. 2022Classifier-Free Diffusion Guidancecited by 5,717
  8. 2022Flow Matching for Generative Modelingcited by 5,426
  9. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,581
  10. 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesiscited by 4,094
  11. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,736
  12. 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flowcited by 3,550
26 more
  1. 2022Elucidating the Design Space of Diffusion-Based Generative Modelscited by 2,994
  2. 2025Wan: Open and Advanced Large-Scale Video Generative Modelscited by 2,536
  3. 2020GLU Variants Improve Transformercited by 1,724
  4. 2021Structured Denoising Diffusion Models in Discrete State-Spacescited by 1,665
  5. 2019Root Mean Square Layer Normalizationcited by 1,425
  6. 2018Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarizationcited by 1,308
  7. 2022Diffusion-LM Improves Controllable Text Generationcited by 1,006
  8. 2022Building Normalizing Flows with Stochastic Interpolantscited by 964
  9. 2025FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Spacecited by 960
  10. 2023Stochastic Interpolants: A Unifying Framework for Flows and Diffusionscited by 897
  11. 2024SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformerscited by 707
  12. 2025Mean Flows for One-step Generative Modelingcited by 524
  13. 2025Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decodingcited by 404
  14. 2022Analog Bits: Generating Discrete Data using Diffusion Models with Self-Conditioningcited by 362
  15. 2020Query-Key Normalization for Transformerscited by 266
  16. 2025MMaDA: Multimodal Large Diffusion Language Modelscited by 241
  17. 2025LLaDA-V: Large Language Diffusion Models with Visual Instruction Tuningcited by 145
  18. 2025Improved Mean Flows: On the Challenges of Fastforward Generative Modelscited by 102
  19. 2024Fisher Flow Matching for Generative Modeling over Discrete Datacited by 68
  20. 2026Categorical Flow Mapscited by 37
  21. 2025Diffusion Models without Classifier-free Guidancecited by 28
  22. 2026Discrete Flow Mapscited by 23
  23. 2025Visual Generation Without Guidancecited by 17
  24. 2025Joint Distillation for Fast Likelihood Evaluation and Sampling in Flow-based Modelscited by 17
  25. 2026Omni-Diffusion: Unified Multimodal Understanding and Generation with Masked Discrete Diffusioncited by 9
  26. 2026Why Gaussian Diffusion Models Fail on Discrete Data and How to Prevent It?cited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 27.

27 papers.

Sort:

By area:

  1. 2026Self-conditioned Flow Map Language Models via Fixed-point Flowscited by 9
  2. 2026Posterior Refinement: Fast Language Generation via Any-Order Flow Mapscited by 8
  3. 2026Flow Reasoning Models: Turning Flows Into Efficient Recurrent Reasonerscited by 5
  4. 2026AURORA-LM: Autoencoding Unified Representation for Continuous-Latent Diffusion Language Modelingcited by 3
  5. 2026Discrete Beckmann Transport Models for One-Step Language Modeling and Reasoningcited by 3
  6. 2026Low Perplexity is Repetition: A One-Dimensional Self-Conditioning Attractor in Continuous Diffusion LMscited by 2
  7. 2026Don't Commit Alone: Joint Token Commitment in Diffusion Language Modelscited by 2
  8. 2026ConvergeFlow: Language Flow with Provable Convergence to Token Embeddingscited by 2
  9. 2026Distilled Continuous Diffusion Language Models Can Write Code in Few Steps---or Onecited by 2
  10. 2026Simplex Diffusion Modelscited by 1
  11. 2026Sticky Jump Diffusions: A Unifying View of Masked, Continuous, and Hybrid Diffusioncited by 0
  12. 2026From Interface to Inference: Eliciting Any-Order Inference from Any-Order Modelscited by 0
15 more
  1. 2026Embedded Graph Flows for Categorical Graph Generationcited by 0
  2. 2026A Ticket from Marginals to Joints: Coupled-Noise Distillation for One-Step Block Generation in Diffusion Language Modelscited by 0
  3. 2026How to Guide Your Language Flowcited by 0
  4. 2026Parallelism, critical windows, and separations among diffusion language modelscited by 0
  5. 2026One Latent, Many Tokens: Jointly Learning Compressed Embeddings for Efficient Language Diffusioncited by 0
  6. 2026KiT: A Foundation Model for Financial Time-Series Forecasting using DiffusionTransformerscited by 0
  7. 2026Reasoning with Continuous Latent Diffusioncited by 0
  8. 2026Fine-Tuning Diffusion Language Models with Context Selection and Target Weightingcited by 0
  9. 2026Distribution Matching Distillation for Continuous Diffusion Language Modelscited by 0
  10. 2026Multimodal Flow: Unified Flow Modeling of Language and Vision in Embedding Spacescited by 0
  11. 2026Gumbel Straight Flow: Distilling Autoregressive Models into One-step Flow Mapscited by 0
  12. 2026Clock Diffusion: Efficient Semi-Autoregressive Continuous Diffusion Language Modelscited by 0
  13. 2026Scaling and Distilling Text Embeddings for Better Diffusibilitycited by 0
  14. 2026Unrolled Flow Models for Reasoningcited by 0
  15. 2026Memorization and Malign Generalization in Conditional Diffusion Models with Random Featurescited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026TextLDM: Language Modeling with Continuous Latent Diffusion
  2. 2026ELF-REG: Scaling Continuous Diffusion Language Models to Reasoning Tasks
  3. 2026Denoising Hierarchical Representations: Joint Continuous Diffusion for Language Modeling
  4. 2025FS-DFM: Fast and Accurate Long Text Generation with Few-Step Diffusion Language Models
  5. 2026Continuous Diffusion Scales Competitively with Discrete Diffusion for Language
  6. 2026dLLM: Simple Diffusion Language Modeling
  7. 2026Continuous Language Diffusion as a Decoder-Interface Problem
  8. 2026Diffusion Language Models: An Experimental Analysis