mlsanityData to 14 Oct 2026

Learning to Parallel: Accelerating Diffusion Large Language Models via Learnable Parallel Decoding

arXiv 2509.25188 ↗First posted 29 Sep 2025cs.CLcited by 27

On the map: Efficient LLM inference and serving → Diffusion LLM and MoE inference

Abstract (arXiv)

Autoregressive decoding in large language models (LLMs) requires O(n) sequential steps for n tokens, fundamentally limiting inference throughput. Recent diffusion-based LLMs (dLLMs) enable parallel token generation through iterative denoising. However, current parallel decoding strategies rely on fixed, input-agnostic heuristics (e.g., confidence thresholds), which fail to adapt to input-specific characteristics, resulting in suboptimal speed-quality trade-offs across diverse NLP tasks. In this work, we explore a more flexible and dynamic approach to parallel decoding. We propose Learning to Parallel Decode (Learn2PD), a framework that trains a lightweight and adaptive filter model to predict, for each token position, whether the current prediction matches the final output. This learned filter approximates an oracle parallel decoding strategy that unmasks tokens only when correctly predicted. Importantly, the filter model is learned in a post-training manner, requiring only a small amount of computation to optimize it (minute-level GPU time). Additionally, we introduce End-of-Text Prediction (EoTP) to detect decoding completion at the end of sequence, avoiding redundant decoding of padding tokens. Experiments on the LLaDA benchmark demonstrate that our method achieves up to 22.58× speedup without any performance drop, and up to 57.51× when combined with KV-Cache.

Key facts. Cited by 27 papers in the map (data through 14 Oct 2026). Placed in Diffusion LLM and MoE inference, one of 12 directions in Efficient LLM inference and serving. Builds on 37 of its 41 references that are in the map. First posted on arXiv 29 Sep 2025.

Builds on

37 of 41 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025Accelerating Diffusion LLMs via Adaptive Parallel Decoding31 May
  2. 2025Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding28 May

Other references · 35

  1. 2017Attention Is All You Needcited by 60,361
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  3. 2020Language Models are Few-Shot Learnerscited by 32,419
  4. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  5. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  6. 2020Denoising Diffusion Implicit Modelscited by 9,781
  7. 2021Evaluating Large Language Models Trained on Codecited by 8,320
  8. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,058
  9. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,761
  10. 2021Program Synthesis with Large Language Modelscited by 3,637
  11. 2023A Survey of Large Language Modelscited by 2,576
  12. 2021Structured Denoising Diffusion Models in Discrete State-Spacescited by 1,719
23 more
  1. 2022Fast Inference from Transformers via Speculative Decodingcited by 1,653
  2. 2021MLP-Mixer: An all-MLP Architecture for Visioncited by 1,446
  3. 2022Solving Quantitative Reasoning Problems with Language Modelscited by 1,392
  4. 2022MaskGIT: Masked Generative Image Transformercited by 1,099
  5. 2022Diffusion-LM Improves Controllable Text Generationcited by 1,027
  6. 2025Large Language Diffusion Modelscited by 912
  7. 2024Simple and Effective Masked Diffusion Language Modelscited by 848
  8. 2024Simplified and Generalized Masked Diffusion for Discrete Datacited by 626
  9. 2021Argmax Flows and Multinomial Diffusion: Learning Categorical Distributionscited by 574
  10. 2025Dream 7B: Diffusion Large Language Modelscited by 544
  11. 2024Large Language Models: A Surveycited by 525
  12. 2022A Continuous Time Framework for Discrete Denoising Modelscited by 451
  13. 2025Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Modelscited by 409
  14. 2024Your Absorbing Discrete Diffusion Secretly Models the Conditional Distributions of Clean Datacited by 381
  15. 2018Blockwise Parallel Decoding for Deep Autoregressive Modelscited by 323
  16. 2024Scaling Diffusion Language Models via Adaptation from Autoregressive Modelscited by 274
  17. 2024Masked Diffusion Models are Secretly Time-Agnostic Masked Models and Exploit Inaccurate Categorical Samplingcited by 208
  18. 2025dLLM-Cache: Accelerating Diffusion Large Language Models with Adaptive Cachingcited by 169
  19. 2025FlashDLM: Accelerating Diffusion Language Model Inference via Efficient KV Caching and Guided Diffusioncited by 77
  20. 2025Diffusion Language Models Know the Answer Before Decodingcited by 70
  21. 2025Accelerating Diffusion Large Language Models with SlowFast Sampling: The Three Golden Principlescited by 45
  22. 2023Diffusion-NAT: Self-Prompting Discrete Diffusion for Non-Autoregressive Text Generationcited by 21
  23. 2023Through the Lens of Core Competency: Survey on Evaluation of Large Language Modelscited by 12

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 27.

27 papers.

Sort:

By area:

  1. 2026DAPD: Dependency-Aware Parallel Decoding via Attention for Diffusion LLMscited by 14
  2. 2026Beyond Hard Masks: Progressive Token Evolution for Diffusion Language Modelscited by 6
  3. 2026LightningRL: Breaking the Accuracy-Parallelism Trade-off of Block-wise dLLMs via Reinforcement Learningcited by 6
  4. 2026FOCUS: DLLMs Know How to Tame Their Compute Boundcited by 3
  5. 2026Locally Coherent Parallel Decoding in Diffusion Language Modelscited by 3
  6. 2026TAD: Temporal-Aware Trajectory Self-Distillation for Fast and Accurate Diffusion LLMcited by 2
  7. 2026TIDE: Efficient and Lossless MoE Diffusion LLM Inference with I/O-aware Expert Offloadcited by 2
  8. 2026Attention-Discounted Adaptive Sampler for Masked Diffusion Language Modelscited by 2
  9. 2026STDec: Spatio-Temporal Stability Guided Decoding for dLLMscited by 1
  10. 2026The Path Matters: Learning a Token-Commitment Policy for Diffusion Language Modelscited by 1
  11. 2026Efficient Diffusion LLMs via Temporal-Spatial Parallel Decoding and Confidence Extrapolationcited by 1
  12. 2026Beyond Fully Random Masking: Attention-Guided Denoising and Optimization for Diffusion Language Modelscited by 1
15 more
  1. 2026Polestar: Drift-Aware Cache Calibration and Token Commitment for Efficient Inference of Diffusion LLMscited by 1
  2. 2026Temporal Self-Distillation: Faster Inference in Discrete Diffusion Language Modelscited by 1
  3. 2026Multi-Token Residual Predictioncited by 0
  4. 2026AsyncLane: Decoupling Refinement from Advancement in Diffusion Language Model Decodingcited by 0
  5. 2026Understanding Evaluation Illusion in Diffusion Large Language Modelscited by 0
  6. 2026DC-Leap: Training-Free Acceleration of dLLMs via Draft-Guided Contiguous Leaping Decodingcited by 0
  7. 2026CORA-Diff: Confidence-Oriented Residual Acceptance for Efficient Diffusion Language Model Inferencecited by 0
  8. 2026Accelerating Diffusion Language Models via Structured Suffix Modelingcited by 0
  9. 2026Prefix-Denoising Consistency: Test-Time Verification for Diffusion Language Modelscited by 0
  10. 2026Dependency-Aware Revocable Decoding for Efficient Diffusion Large Language Model Inferencecited by 0
  11. 2026Visual Information-Guided Parallel Decoding for Diffusion Multimodal Large Language Modelscited by 0
  12. 2026Diffusion Language Models for Mobile Edge Agentic AI: Foundations, Applications, and Challengescited by 0
  13. 2026Early-Bird Decoding: Accelerating Diffusion LLMs with Learnable Block Sizes and Parallel Samplingcited by 0
  14. 2026d-OPD: Future-Aware On-Policy Distillation for Block Diffusion Language Modelscited by 0
  15. 2026Batch-Paced: Fixed-Budget Batch Decoding in Discrete Diffusion Language Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Why Diffusion Language Models Struggle with Truly Parallel (Non-Autoregressive) Decoding?
  2. 2026Roll Out and Roll Back: Diffusion LLMs are Their Own Efficiency Teachers
  3. 2025Free Draft-and-Verification: Toward Lossless Parallel Decoding for Diffusion Large Language Models
  4. 2026Divide and Conquer: Accelerating Diffusion-Based Large Language Models via Adaptive Parallel Decoding
  5. 2026LEAP: Unlocking dLLM Parallelism via Lookahead Early-Convergence Token Detection
  6. 2026Mean-Field Parallel Decoding for Discrete Diffusion Language Models
  7. 2025Beyond Next-Token Prediction: A Performance Characterization of Diffusion versus Autoregressive Language Models
  8. 2025From Bits to Rounds: Parallel Decoding with Exploration for Diffusion Language Models