mlsanityData to 14 Oct 2026

VoidPadding: Let [VOID] Handle Padding in Masked Diffusion Language Models so that [EOS] Can Focus on Semantic Termination

arXiv 2606.17999 ↗First posted 16 Jun 2026cs.CLcited by 1

On the map: Diffusion models → Discrete and text diffusion

Abstract (arXiv)

MDLMs generate text by denoising a preallocated masked response canvas, making response-length modeling central to instruction tuning. Existing MDLMs often inherit the autoregressive convention of using repeated texttt[EOS] tokens for padding during instruction tuning, giving texttt[EOS] a dual role as both a semantic terminator and a padding token. We show that this dual role is a root cause of texttt[EOS] overflow under large-block decoding. To decouple these roles, we propose VoidPadding, which introduces texttt[VOID] for padding and reserves texttt[EOS] for termination. During inference, the learned texttt[EOS] signal enables early stopping, while the learned texttt[VOID] signal guides adaptive response canvas expansion. On Dream-7B-Instruct, VoidPadding improves the block-size-averaged four-task mean across mathematical reasoning and code generation benchmarks by (+17.84) points over the original model and (+6.95) points over RainbowPadding, while reducing decoding NFE by 55.7% on average. Code is available at https://github.com/Haru-LCY/VoidPadding.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Discrete and text diffusion, one of 14 directions in Diffusion models. Builds on 31 of its 47 references that are in the map. First posted on arXiv 16 Jun 2026.

Builds on

31 of 47 references.

Closest ancestors · 19

Same direction, newest first

  1. 2026Diffusion Language Models Are Natively Length-Aware6 Mar
  2. 2026Improving Sampling for Masked Diffusion Models via Information Gain20 Feb
  3. 2026Improving Variable-Length Generation in Diffusion Language Models via Length Regularization7 Feb
  4. 2026ρ-textttEOS: Training-free Bidirectional Variable-Length Control for Masked Diffusion LLMs30 Jan
  5. 2025WeDLM: Reconciling Diffusion Language Models with Standard Causal Attention for Fast Inference28 Dec
  6. 2025LLaDA2.0: Scaling Up Diffusion Language Models to 100B10 Dec
  7. 2025Soft-Masked Diffusion Language Models20 Oct
  8. 2025Any-Order Flexible Length Masked Diffusion31 Aug
  9. 2025Diffusion Language Models Know the Answer Before Decoding27 Aug
  10. 2025The Diffusion Duality12 Jun
  11. 2025DLM-One: Diffusion Language Models for One-Step Sequence Generation30 May
  12. 2025FlashDLM: Accelerating Diffusion Language Model Inference via Efficient KV Caching and Guided Diffusion27 May
7 more
  1. 2025Continuous Diffusion Model for Language Modeling17 Feb
  2. 2025Train for the Worst, Plan for the Best: Understanding Token Ordering in Masked Diffusions10 Feb
  3. 2024Discrete Diffusion Language Model for Efficient Text Summarization25 Jun
  4. 2024Simplified and Generalized Masked Diffusion for Discrete Data6 Jun
  5. 2023Discrete Diffusion Modeling by Estimating the Ratios of the Data Distribution25 Oct
  6. 2022DiffusionBERT: Improving Generative Masked Language Models with Diffusion Models28 Nov
  7. 2022SSD-LM: Semi-autoregressive Simplex-based Diffusion Language Model for Text Generation and Modular Control31 Oct

Other references · 12

  1. 2017Attention Is All You Needcited by 60,361
  2. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,371
  3. 2021Structured Denoising Diffusion Models in Discrete State-Spacescited by 1,719
  4. 2021Argmax Flows and Multinomial Diffusion: Learning Categorical Distributionscited by 574
  5. 2025Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decodingcited by 420
  6. 2025Accelerating Diffusion LLMs via Adaptive Parallel Decodingcited by 112
  7. 2025Sparse-dLLM: Accelerating Diffusion LLMs with Dynamic Cache Evictioncited by 65
  8. 2025Accelerating Diffusion Large Language Models with SlowFast Sampling: The Three Golden Principlescited by 45
  9. 2025Diffusion LLM with Native Variable Generation Lengths: Let [EOS] Lead the Waycited by 22
  10. 2025Rainbow Padding: Mitigating Early Termination in Instruction-Tuned Diffusion LLMscited by 10
  11. 2026DSB: Dynamic Sliding Block Scheduling for Diffusion LLMscited by 7
  12. 2026Dynamic-dLLM: Dynamic Cache-Budget and Adaptive Parallel Decoding for Training-Free Acceleration of Diffusion LLMcited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Answer First, Reason Later: When Commitment Order Costs Accuracy in Diffusion Language Modelscited by 3

Similar, not cited

Close in wording, no citation either way

  1. 2026Supportive Token Revealing for Fast Diffusion Language Model Decoding
  2. 2026Parallelism and Generation Order in Masked Diffusion Language Models: Limits Today, Potential Tomorrow
  3. 2025Activation Steering for Masked Diffusion Language Models
  4. 2026CARVE: Verified Expansion for Variable-Length Generation in Diffusion Language Models
  5. 2026Register Tokens for Bounded-State Reasoning in Diffusion Language Models
  6. 2026AsyncLane: Decoupling Refinement from Advancement in Diffusion Language Model Decoding
  7. 2026TRIMS: Trajectory-Ranked Instruction Masked Supervision for Diffusion Language Models
  8. 2026CAI-DLLM: Convergence Aware Inference for Diffusion Language Models