Flashdlm: Accelerating diffusion language model inference via efficient kv caching and guided diffusion.arXiv preprint arXiv:2505.21467,

representative citing papers

citing papers explorer

Showing 17 of 17 citing papers.

• Keep The Essentials: Efficient Reference Conditioned Generation via Token Dropping cs.CV · 2026-06-22 · unverdicted · none · ref 8

  Sparse Context achieves 2-4x faster inference in reference-conditioned diffusion models by fine-tuning with random token dropping and applying task-aware selection at inference time, without loss of visual quality.

• AsyncLane: Decoupling Refinement from Advancement in Diffusion Language Model Decoding cs.CL · 2026-06-07 · unverdicted · none · ref 5

  AsyncLane decouples refinement from advancement in DLM decoding via lane forking at delimiters plus efficiency optimizations, yielding up to 3x throughput gains on math and code benchmarks without retraining.

• Self-Augmenting Retrieval for Diffusion Language Models cs.CL · 2026-06-04 · unverdicted · none · ref 43

  SARDI uses lookahead tokens from low-confidence predictions in discrete diffusion language models to dynamically guide retrieval during denoising, outperforming training-free baselines on five multi-hop QA benchmarks at up to 8x higher throughput.

• DMax: Aggressive Parallel Decoding for dLLMs cs.LG · 2026-04-09 · conditional · none · ref 35 · 2 links

  DMax uses On-Policy Uniform Training and Soft Parallel Decoding to enable aggressive parallelism in dLLMs, raising TPF on GSM8K from 2.04 to 5.47 and on MBPP from 2.71 to 5.86 while preserving accuracy.

• Locally Coherent Parallel Decoding in Diffusion Language Models cs.CL · 2026-03-03 · unverdicted · none · ref 6

  CoDiLA adds a compact auxiliary AR model on diffusion latents to enforce local sequential validity during parallel token sampling in discrete diffusion language models.

• Not All Denoising Steps Are Equal: Model Scheduling for Faster Masked Diffusion Language Models cs.LG · 2026-02-04 · unverdicted · none · ref 13

  Early and late denoising steps in masked diffusion LMs are robust to smaller-model replacement, enabling 17% FLOPs reduction with modest generative quality loss.

• VoidPadding: Let [VOID] Handle Padding in Masked Diffusion Language Models so that [EOS] Can Focus on Semantic Termination cs.CL · 2026-06-16 · unverdicted · none · ref 19

  VoidPadding decouples padding from termination in MDLMs via a new [VOID] token, delivering +17.84 average benchmark points and 55.7% fewer decoding steps on Dream-7B-Instruct.

• Efficient Diffusion LLMs via Temporal-Spatial Parallel Decoding and Confidence Extrapolation cs.CL · 2026-05-29 · unverdicted · none · ref 7

  Introduces TSPD with a trajectory-feature controller and training-free CE to reduce denoising steps in dLLMs while aiming to preserve quality.

• BlockBatch: Multi-Scale Consensus Decoding for Efficient Diffusion Language Model Inference cs.LG · 2026-05-28 · unverdicted · none · ref 5

  BlockBatch is a training-free framework that coordinates multiple block-size branches via token merging and synchronization to reduce denoising NFEs by 26.6% and achieve 1.33x speedup in dLLM inference.

• PulseCol: Periodically Refreshed Column-Sparse Attention for Accelerating Diffusion Language Models cs.CL · 2026-05-20 · unverdicted · none · ref 11

  PulseCol introduces periodically refreshed column-sparse attention to achieve up to 1.95x speedup over FlashAttention in diffusion LLMs with maintained model quality.

• Elastic-dLLM: Position Preserving Context Compression and Augmentation of Diffusion LLMs cs.LG · 2026-05-18 · unverdicted · none · ref 9

  Position-preserving MASK token compression reduces redundancy in diffusion LLMs to accelerate parallel decoding and enable context folding for longer sequences.

• $R^2$-dLLM: Accelerating Diffusion Large Language Models via Spatio-Temporal Redundancy Reduction cs.CL · 2026-04-21 · conditional · none · ref 8

  Voltage pulses create nonvolatile, multi-state metastable CDW phases in bulk EuTe4 at room temperature via out-of-plane phase switching in its moiré superstructure.

• STDec: Spatio-Temporal Stability Guided Decoding for dLLMs cs.CL · 2026-04-07 · unverdicted · none · ref 3

  STDec raises dLLM decoding speed by up to 14x on benchmarks like MBPP by using observed spatio-temporal stability to create dynamic, token-specific confidence thresholds while preserving task performance.

• ETS: Energy-Guided Test-Time Scaling for Training-Free RL Alignment cs.LG · 2026-01-29 · unverdicted · none · ref 13 · 2 links

  ETS performs training-free RL alignment for language models by energy-guided test-time scaling with Monte Carlo energy estimation and importance sampling acceleration.

• Diffusion Language Models Know the Answer Before Decoding cs.CL · 2025-08-27 · conditional · none · ref 9

  DLMs show early answer convergence allowing Prophet to cut decoding steps by up to 3.4x on LLaDA-8B and Dream-7B while keeping output quality.

• Understanding Evaluation Illusion in Diffusion Large Language Models cs.CL · 2026-06-28 · unverdicted · none · ref 10 · 2 links

  Prompt template choice strongly affects apparent performance of parallel decoding methods in dLLMs, causing inconsistent rankings and illusory gains; current methods fail to beat single-token decoding.

• SimSD: Simple Speculative Decoding in Diffusion Language Models cs.CL · 2026-06-01 · unreviewed · ref 25