mlsanityData to 7 Oct 2026

Diffusion Guided Language Modeling

arXiv 2408.04220 ↗First posted 8 Aug 2024cs.CL cs.LGcited by 15

On the map: Diffusion models → Discrete and text diffusion

Abstract (arXiv)

Current language models demonstrate remarkable proficiency in text generation. However, for many applications it is desirable to control attributes, such as sentiment, or toxicity, of the generated language -- ideally tailored towards each specific use case and target audience. For auto-regressive language models, existing guidance methods are prone to decoding errors that cascade during generation and degrade performance. In contrast, text diffusion models can easily be guided with, for example, a simple linear sentiment classifier -- however they do suffer from significantly higher perplexity than auto-regressive alternatives. In this paper we use a guided diffusion model to produce a latent proposal that steers an auto-regressive language model to generate text with desired properties. Our model inherits the unmatched fluency of the auto-regressive approach and the plug-and-play flexibility of diffusion. We show that it outperforms previous plug-and-play guidance methods across a wide range of benchmark data sets. Further, controlling a new attribute in our framework is reduced to training a single logistic regression classifier.

Key facts. Cited by 15 papers in the map (data through 7 Oct 2026). Placed in Discrete and text diffusion, one of 14 directions in Diffusion models. Builds on 47 of its 58 references that are in the map. First posted on arXiv 8 Aug 2024.

Builds on

47 of 58 references.

Closest ancestors · 3

Same direction, newest first

  1. 2023PLANNER: Generating Diversified Paragraph via Latent Language Diffusion Model5 Jun
  2. 2023Likelihood-Based Diffusion Language Models30 May
  3. 2022DiffuSeq: Sequence to Sequence Text Generation with Diffusion Models17 Oct

Other references · 44

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  4. 2022Training language models to follow instructions with human feedbackcited by 15,712
  5. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  6. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,266
  7. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  8. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,185
  9. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  10. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,322
  11. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,462
  12. 2019DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lightercited by 4,243
32 more
  1. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,067
  2. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,736
  3. 2022Elucidating the Design Space of Diffusion-Based Generative Modelscited by 2,994
  4. 2023Sparks of Artificial General Intelligence: Early experiments with GPT-4cited by 2,579
  5. 2020GLU Variants Improve Transformercited by 1,724
  6. 2020Don't Stop Pretraining: Adapt Language Models to Domains and Taskscited by 1,548
  7. 2019Root Mean Square Layer Normalizationcited by 1,425
  8. 2022Diffusion Posterior Sampling for General Noisy Inverse Problemscited by 1,375
  9. 2020RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Modelscited by 1,235
  10. 2021Cascaded Diffusion Models for High Fidelity Image Generationcited by 1,155
  11. 2021Variational Diffusion Modelscited by 1,145
  12. 2022Diffusion-LM Improves Controllable Text Generationcited by 1,006
  13. 2022eDiff-I: Text-to-Image Diffusion Models with an Ensemble of Expert Denoiserscited by 921
  14. 2020On Layer Normalization in the Transformer Architecturecited by 906
  15. 2019Plug and Play Language Models: A Simple Approach to Controlled Text Generationcited by 793
  16. 2023Scaling Vision Transformers to 22 Billion Parameterscited by 690
  17. 2021MAUVE: Measuring the Gap Between Neural Text and Human Text using Divergence Frontierscited by 466
  18. 2021Sentence-T5: Scalable Sentence Encoders from Pre-trained Text-to-Text Modelscited by 462
  19. 2021DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Expertscited by 434
  20. 2023Fine-Grained Human Feedback Gives Better Rewards for Language Model Trainingcited by 399
  21. 2020GeDi: Generative Discriminator Guided Sequence Generationcited by 377
  22. 2021FUDGE: Controlled Text Generation With Future Discriminatorscited by 374
  23. 2023Understanding Diffusion Objectives as the ELBO with Simple Data Augmentationcited by 293
  24. 2023Personalized Soups: Personalized Large Language Model Alignment via Post-hoc Parameter Mergingcited by 271
  25. 2022A Contrastive Framework for Neural Text Generationcited by 224
  26. 2023Text Embeddings Reveal (Almost) As Much As Textcited by 189
  27. 2022Cramming: Training a Language Model on a Single GPU in One Daycited by 81
  28. 2023Reward-Augmented Decoding: Efficient Controlled Text Generation With a Unidirectional Reward Modelcited by 79
  29. 2020Multi-agent Communication meets Natural Language: Synergies between Functional and Structural Language Learningcited by 65
  30. 2022On the Blind Spots of Model-Based Evaluation Metrics for Text Generationcited by 48
  31. 2023On the Challenges of Using Black-Box APIs for Toxicity Evaluation in Researchcited by 44
  32. 2023Language Model Alignment with Elastic Resetcited by 34

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 2; 2025: 7; 2026: 6.

15 papers.

Sort:

By area:

  1. 2024Scaling up Masked Diffusion Models on Textcited by 195
  2. 2024Large Concept Models: Language Modeling in a Sentence Representation Spacecited by 86
  3. 2025Theoretical Benefit and Limitation of Diffusion Language Modelcited by 46
  4. 2026ELF: Embedded Language Flowscited by 27
  5. 2025LaDiR: Latent Diffusion Enhances LLMs for Text Reasoningcited by 25
  6. 2025Adaptive Classifier-Free Guidance via Dynamic Low-Confidence Maskingcited by 21
  7. 2026Learning a Generative Meta-Model of LLM Activationscited by 8
  8. 2025Language Models Are Implicitly Continuouscited by 4
  9. 2025From Text to Talk: Audio-Language Model Needs Non-Autoregressive Joint Trainingcited by 3
  10. 2025Simple Denoising Diffusion Language Modelscited by 2
  11. 2026Stop-Think-AutoRegress: Language Modeling with Latent Diffusion Planningcited by 2
  12. 2026HIVE: Hidden-Evidence Verification for Hallucination Detection in Diffusion Large Language Modelscited by 1
3 more
  1. 2025Next-Token Prediction Should be Ambiguity-Sensitive: A Meta-Learning Perspectivecited by 0
  2. 2026Focus on the Core: Empowering Diffusion Large Language Models by Self-Contrastcited by 0
  3. 2026Repurposing Pre-trained LLMs as High Fidelity Continuous Text Autoencoderscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Scaling Diffusion Language Models via Adaptation from Autoregressive Models
  2. 2023A Survey of Diffusion Models in Natural Language Processing
  3. 2023Diffusion Models for Non-autoregressive Text Generation: A Survey
  4. 2026Steering Without Breaking: Mechanistically Informed Interventions for Discrete Diffusion Language Models
  5. 2023AR-Diffusion: Auto-Regressive Diffusion Model for Text Generation
  6. 2024Language Rectified Flow: Advancing Diffusion Language Generation with Probabilistic Flows
  7. 2026Diffusion Language Models: An Experimental Analysis
  8. 2022Latent Diffusion for Language Generation