mlsanityData to 7 Oct 2026

DAWN: Dependency-Aware Fast Inference for Diffusion LLMs

arXiv 2602.06953 ↗First posted 6 Feb 2026cs.CLcited by 12

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

Abstract (arXiv)

Diffusion large language models (dLLMs) have shown advantages in text generation, particularly due to their inherent ability for parallel decoding. However, constrained by the quality--speed trade-off, existing inference solutions adopt conservative parallel strategies, leaving substantial efficiency potential underexplored. A core challenge is that parallel decoding assumes each position can be filled independently, but tokens are often semantically coupled. Thus, the correct choice at one position constrains valid choices at others. Without modeling these inter-token dependencies, parallel strategies produce deteriorated outputs. Motivated by this insight, we propose DAWN, a training-free, dependency-aware decoding method for fast dLLM inference. DAWN extracts token dependencies and leverages two key motivations: (1) positions dependent on unmasked certain positions become more reliable, (2) simultaneously unmasking strongly coupled uncertain positions induces errors. Given those findings, DAWN leverages a dependency graph to select more reliable unmasking positions at each iteration, achieving high parallelism with negligible loss in generation quality. Extensive experiments across multiple models and datasets demonstrate that DAWN speedups the inference by 1.80-8.06x over baselines while preserving the generation quality. Code is released at https://github.com/lizhuo-luo/DAWN.

Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Placed in Diffusion LLM and MoE inference, one of 12 directions in Efficient LLM inference and serving. Builds on 41 of its 46 references that are in the map. First posted on arXiv 6 Feb 2026.

Builds on

41 of 46 references.

Closest ancestors · 8

Same direction, newest first

  1. 2026d3LLM: Ultra-Fast Diffusion LLM using Pseudo-Trajectory Distillation12 Jan
  2. 2025Orchestrating Dual-Boundaries: An Arithmetic Intensity Inspired Acceleration Framework for Diffusion Language Models24 Nov
  3. 2025Accelerating Diffusion LLM Inference via Local Determinism Propagation8 Oct
  4. 2025dParallel: Learnable Parallel Decoding for dLLMs30 Sep
  5. 2025Structuring The Future: Diffusion LLM Speculative Decoding via Calibrated Draft Graphs22 Sep
  6. 2025Sparse-dLLM: Accelerating Diffusion LLMs with Dynamic Cache Eviction4 Aug
  7. 2025Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding28 May
  8. 2025dKV-Cache: The Cache for Diffusion Language Models21 May

Other references · 33

  1. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  2. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  3. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,203
  4. 2023SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesiscited by 4,290
  5. 2021Program Synthesis with Large Language Modelscited by 3,509
  6. 2023Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasetscited by 2,526
  7. 2023Efficient Streaming Language Models with Attention Sinkscited by 2,115
  8. 2024HunyuanVideo: A Systematic Framework For Large Video Generative Modelscited by 1,597
  9. 2025Large Language Diffusion Modelscited by 872
  10. 2024An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Modelscited by 629
  11. 2024Simplified and Generalized Masked Diffusion for Discrete Datacited by 606
  12. 2025Dream 7B: Diffusion Large Language Modelscited by 515
21 more
  1. 2023The Reversal Curse: LLMs trained on "A is B" fail to learn "B is A"cited by 399
  2. 2024SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformerscited by 331
  3. 2025EAGLE-3: Scaling up Inference Acceleration of Large Language Models via Training-Time Testcited by 318
  4. 2025LLaDA 1.5: Variance-Reduced Preference Optimization for Large Language Diffusion Modelscited by 273
  5. 2025MMaDA: Multimodal Large Diffusion Language Modelscited by 241
  6. 2025Mercury: Ultra-Fast Language Models Based on Diffusioncited by 225
  7. 2025Sparse VideoGen: Accelerating Video Diffusion Transformers with Spatial-Temporal Sparsitycited by 204
  8. 2025Seed Diffusion: A Large-Scale Diffusion Language Model with High-Speed Inferencecited by 173
  9. 2025dLLM-Cache: Accelerating Diffusion Large Language Models with Adaptive Cachingcited by 168
  10. 2024SageAttention: Accurate 8-Bit Attention for Plug-and-play Inference Accelerationcited by 158
  11. 2025LLaDA-V: Large Language Diffusion Models with Visual Instruction Tuningcited by 145
  12. 2025Accelerated Sampling from Masked Diffusion Models via Entropy Bounded Unmaskingcited by 128
  13. 2025Fast-dLLM v2: Efficient Block-Diffusion LLMcited by 128
  14. 2025Wide-In, Narrow-Out: Revokable Decoding for Efficient and Effective DLLMscited by 45
  15. 2025KLASS: KL-Guided Fast Inference in Masked Diffusion Modelscited by 44
  16. 2025WeDLM: Reconciling Diffusion Language Models with Standard Causal Attention for Fast Inferencecited by 38
  17. 2025Beyond Fixed: Training-Free Variable-Length Denoising for Diffusion Large Language Modelscited by 32
  18. 2025Diffusion LLM with Native Variable Generation Lengths: Let [EOS] Lead the Waycited by 22
  19. 2025Attention Sinks in Diffusion Language Modelscited by 18
  20. 2025Ideas in Inference-time Scaling can Benefit Generative Pre-training Algorithmscited by 12
  21. 2025Quant-dLLM: Post-Training Extreme Low-Bit Quantization for Diffusion Large Language Modelscited by 11

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 12.

12 papers.

  1. 2026DAPD: Dependency-Aware Parallel Decoding via Attention for Diffusion LLMscited by 13
  2. 2026R2-dLLM: Accelerating Diffusion Large Language Models via Spatio-Temporal Redundancy Reductioncited by 6
  3. 2026Attention-Discounted Adaptive Sampler for Masked Diffusion Language Modelscited by 2
  4. 2026DepCap: Adaptive Block-Wise Parallel Decoding for Efficient Diffusion LM Inferencecited by 1
  5. 2026Polestar: Drift-Aware Cache Calibration and Token Commitment for Efficient Inference of Diffusion LLMscited by 1
  6. 2026Ripple-Pivot Search: Active Parallel Decoding for Diffusion Large Language Modelscited by 1
  7. 2026Unified Energy for Invariant and Independent Decoding in Diffusion Language Modelscited by 0
  8. 2026Mean-Field Parallel Decoding for Discrete Diffusion Language Modelscited by 0
  9. 2026CORA-Diff: Confidence-Oriented Residual Acceptance for Efficient Diffusion Language Model Inferencecited by 0
  10. 2026Prefix-Denoising Consistency: Test-Time Verification for Diffusion Language Modelscited by 0
  11. 2026Affix Cache for Diffusion Large Language Modelscited by 0
  12. 2026Fine-Tuning Diffusion Language Models with Context Selection and Target Weightingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Dependency-Guided Parallel Decoding in Discrete Diffusion Language Models
  2. 2026Accelerating Masked Diffusion Large Language Models: A Survey of Efficient Inference Techniques
  3. 2026Streaming-dLLM: Accelerating Diffusion LLMs via Suffix Pruning and Dynamic Decoding
  4. 2026Early-Bird Decoding: Accelerating Diffusion LLMs with Learnable Block Sizes and Parallel Sampling
  5. 2026Fast-dLLM++: Fr'echet Profile Decoding for Faster Diffusion LLM Inference
  6. 2025Free Draft-and-Verification: Toward Lossless Parallel Decoding for Diffusion Large Language Models
  7. 2025Diffusion Language Model Inference with Monte Carlo Tree Search
  8. 2026Roll Out and Roll Back: Diffusion LLMs are Their Own Efficiency Teachers