mlsanityData to 14 Oct 2026

Training-Free Self-Correction for Multimodal Masked Diffusion Models

arXiv 2602.02927 ↗First posted 2 Feb 2026stat.ML cs.LGcited by 2

On the map: Diffusion models → Discrete and text diffusion

Abstract (arXiv)

Masked diffusion models have emerged as a powerful framework for text and multimodal generation. However, their sampling procedure updates multiple tokens simultaneously and treats generated tokens as immutable, which may lead to error accumulation when early mistakes cannot be revised. In this work, we revisit existing self-correction methods and identify limitations stemming from additional training requirements or reliance on misaligned likelihood estimates. We propose a training-free self-correction framework that exploits the inductive biases of pre-trained masked diffusion models. Without modifying model parameters or introducing auxiliary evaluators, our method significantly improves generation quality on text-to-image generation and multimodal understanding tasks with reduced sampling steps. Moreover, the proposed framework generalizes across different masked diffusion architectures, highlighting its robustness and practical applicability. Code can be found in https://github.com/huge123/FreeCorrection.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Discrete and text diffusion, one of 14 directions in Diffusion models. Builds on 34 of its 41 references that are in the map. First posted on arXiv 2 Feb 2026.

Builds on

34 of 41 references.

Closest ancestors · 18

Same direction, newest first

  1. 2025KLASS: KL-Guided Fast Inference in Masked Diffusion Models7 Nov
  2. 2025SDAR: A Synergistic Diffusion-AutoRegression Paradigm for Scalable Sequence Generation7 Oct
  3. 2025Fine-Tuning Masked Diffusion for Provable Self-Correction1 Oct
  4. 2025Don't Settle Too Early: Self-Reflective Remasking for Diffusion Language Models28 Sep
  5. 2025Revolutionizing Reinforcement Learning Framework for Diffusion Large Language Models8 Sep
  6. 2025Dream 7B: Diffusion Large Language Models21 Aug
  7. 2025Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models12 Mar
  8. 2025Generalized Interpolating Discrete Diffusion6 Mar
  9. 2025Remasking Discrete Diffusion Models with Inference-Time Scaling1 Mar
  10. 2025Large Language Diffusion Models14 Feb
  11. 2025Train for the Worst, Plan for the Best: Understanding Token Ordering in Masked Diffusions10 Feb
  12. 2025Path Planning for Masked Diffusion Model Sampling5 Feb
6 more
  1. 2024Beyond Autoregression: Discrete Diffusion for Complex Reasoning and Planning18 Oct
  2. 2024Think While You Generate: Discrete Diffusion with Planned Denoising8 Oct
  3. 2024Simplified and Generalized Masked Diffusion for Discrete Data6 Jun
  4. 2024Your Absorbing Discrete Diffusion Secretly Models the Conditional Distributions of Clean Data6 Jun
  5. 2023Discrete Diffusion Modeling by Estimating the Ratios of the Data Distribution25 Oct
  6. 2023A Reparameterized Discrete Diffusion Model for Text Generation11 Feb

Other references · 16

  1. 2017Neural Discrete Representation Learningcited by 4,712
  2. 2023MMBench: Is Your Multi-modal Model an All-around Player?cited by 2,141
  3. 2023MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGIcited by 2,080
  4. 2021Structured Denoising Diffusion Models in Discrete State-Spacescited by 1,719
  5. 2023GenEval: An Object-Focused Framework for Evaluating Text-to-Image Alignmentcited by 972
  6. 2023SEED-Bench: Benchmarking Multimodal LLMs with Generative Comprehensioncited by 940
  7. 2024VLMEvalKit: An Open-Source Toolkit for Evaluating Large Multi-Modality Modelscited by 455
  8. 2022A Continuous Time Framework for Discrete Denoising Modelscited by 451
  9. 2024Discrete Flow Matchingcited by 317
  10. 2024Generative Flows on Discrete State-Spaces: Enabling Multimodal Flows with Applications to Protein Co-Designcited by 305
  11. 2025MMaDA: Multimodal Large Diffusion Language Modelscited by 250
  12. 2022Score-based Continuous-time Discrete Diffusion Modelscited by 169
4 more
  1. 2024Diffusion Language Models Are Versatile Protein Learnerscited by 98
  2. 2025Lumina-DiMOO: An Omni Diffusion Large Language Model for Multi-Modal Generation and Understandingcited by 88
  3. 2024Informed Correctors for Discrete Diffusion Modelscited by 57
  4. 2024textitJump Your Steps: Optimizing Sampling Schedule of Discrete Diffusion Modelscited by 40

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Concurrent Image Understanding and Generation: Self-Correcting Coupled Markov Jump Processescited by 1
  2. 2026Entropy Can Flow, or It Can Guide. Be Entropy. LEDFlow: Introducing Entropy-guided Generation Order into Uniform Discrete Flowcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Learn from Your Mistakes: Self-Correcting Masked Diffusion Models
  2. 2024Simple and Effective Masked Diffusion Language Models
  3. 2026The Design Space of Tri-Modal Masked Diffusion Models
  4. 2026Remask, Don't Replace: Token-to-Mask Refinement in Diffusion Language Models
  5. 2025Dimathtt[M]O: Distilling Masked Diffusion Models into One-step Generator
  6. 2025Sparse-LaViDa: Sparse Multimodal Discrete Diffusion Language Models
  7. 2026Revise, Don't Freeze: Sampler-Matched Training for Self-Correcting Masked Diffusion Language Models
  8. 2025Multimodal LLM-Guided Semantic Correction in Text-to-Image Diffusion