arXiv:2602. 11133v2 Announce Type: replace Abstract: Diffusion language models generate text through iterative refinement, a process that is often computationally inefficient because many tokens reach stability long before the final denoising step.
arXiv:2609.38066v1 Announce Type: new Abstract: Discrete diffusion language models can generate multiple tokens in parallel, but reducing the number of denoising steps can lead to inconsistent predic...
arXiv:2601. 07568v3 Announce Type: replace-cross Abstract: Diffusion large language models (dLLMs) offer capabilities beyond those of autoregressive (AR) LLMs, such as parallel decoding and random-order generation.
Diffusion Language Models (DLMs) exhibit strong parallel decoding capabilities by denoising multiple tokens in a single generation step. However, this parallelism comes with substantial computational...
The paper introduces a training‑free structured suffix modeling technique to accelerate Diffusion Language Models (DLMs). It partitions the suffix into local, middle, and tail regions, retaining varying numbers of tokens per region and incorporating previous decoding results into current token representations. Experiments on three DLMs show significant speedups—up to 72.81× in long‑sequence inference—while often improving performance, and the method is compatible with existing acceleration strategies.
arXiv:2508. 10875v3 Announce Type: replace-cross Abstract: Diffusion Language Models (DLMs) are rapidly emerging as a powerful and promising alternative to the dominant autoregressive (AR) paradigm.
The article "How To Build A Modern Diffusion Language Model" reviews the recent advances in large language models (LLMs) and highlights their limitations, such as sequential token generation that underutilizes GPU hardware and prevents error correction. It explains how discrete diffusion models overcome these issues by generating tokens in parallel through iterative denoising, enabling high-speed inference and controllable generation. The paper surveys key concepts, architectural components, and applications of diffusion LLMs in biology and language, and discusses their implications for future scaling laws.
Diffusion language models accelerate generation by decoding multiple masked positions per forward pass. However, accuracy is known to drop when parallel decoding trajectories deviate from their sequen...
The paper introduces dSpark, a schedule distillation framework that learns when to commit tokens in diffusion large language models to accelerate generation. By constructing unmasking schedules that prioritize revealing tokens which make other positions easier to predict, dSpark reduces uncertainty earlier in decoding. Experiments on mathematical reasoning and code generation tasks show significant speedups—10.12× on LLaDA‑8B‑Instruct and 5.12× on Dream‑7B‑Instruct—while preserving superior task performance.
arXiv:2606. 19475v1 Announce Type: new Abstract: Large Language Models (LLMs) have revolutionized language modeling through autoregressive generation, enabling strong performance across a wide range of tasks.
Diffusion large language models (DLLMs) enable non-autoregressive generation by iteratively denoising corrupted token sequences with bidirectional context. Despite their ability to update multiple positions in parallel, inference remains costly due to the many denoising steps required for high-quality generation.
TeCoP is a training‑free decoder that accelerates diffusion large language model (dLLM) decoding by decoupling prediction readiness from context utility. It uses tentative context pipelining and draft‑and‑verify to reuse stable but uncommitted predictions as provisional cross‑block context, with commitment based on confidence and stability checks. Across four dLLMs on GSM8K, Minerva MATH, and MBPP, TeCoP cuts forward evaluations by up to 87% and boosts throughput up to 7.5‑fold without KV caching and 11.77‑fold with caching, outperforming Fast‑dLLM and DSB while maintaining accuracy within one percentage point.