FS-DFM: Fast and Accurate Long Text Generation with Few-Step Diffusion Language Models

• URL: http://arxiv.org/abs/2509.20624v1
• Date: Wed, 24 Sep 2025 23:59:05 GMT
• Title: FS-DFM: Fast and Accurate Long Text Generation with Few-Step Diffusion Language Models
• Abstract summary: We introduce FS-DFM, Few-Step Discrete Flow-Matching.<n>A discrete flow-matching model designed for speed without sacrificing quality.<n>On language modeling benchmarks, FS-DFM with 8 sampling steps achieves perplexity parity with a 1,024-step discrete-flow baseline.
• Score: 6.463102424844334
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Autoregressive language models (ARMs) deliver strong likelihoods, but are inherently serial: they generate one token per forward pass, which limits throughput and inflates latency for long sequences. Diffusion Language Models (DLMs) parallelize across positions and thus appear promising for language generation, yet standard discrete diffusion typically needs hundreds to thousands of model evaluations to reach high quality, trading serial depth for iterative breadth. We introduce FS-DFM, Few-Step Discrete Flow-Matching. A discrete flow-matching model designed for speed without sacrificing quality. The core idea is simple: make the number of sampling steps an explicit parameter and train the model to be consistent across step budgets, so one big move lands where many small moves would. We pair this with a reliable update rule that moves probability in the right direction without overshooting, and with strong teacher guidance distilled from long-run trajectories. Together, these choices make few-step sampling stable, accurate, and easy to control. On language modeling benchmarks, FS-DFM with 8 sampling steps achieves perplexity parity with a 1,024-step discrete-flow baseline for generating 1,024 tokens using a similar-size model, delivering up to 128 times faster sampling and corresponding latency/throughput gains.

Related papers

• Distilled Continuous Diffusion Language Models Can Write Code in Few Steps---or One [16.38284188874483]
  We introduce PlaidQ, a 0.7B continuous diffusion language model for code generation.<n>We show that PlaidQ's trajectory can be aggressively distilled into only a few denoising steps.<n>Together, these results establish continuous diffusion as a viable path to few-step and one-step code generation.
  arXiv  Detail & Related papers  (2026-09-03T22:42:57Z)
• Masked Language Flow Models [10.204151640924454]
  Flow Language Models (FLMs) sidestep this limitation by learning a continuous flow that transports noise toward clean sequences.<n>We introduce Masked Language Flow Models (MLFMs) which incorporate masking into FLMs using a continuous interpolant to bridge partially masked and clean sequences.<n>We find, for the first time, that flow-based language models can be scaled to solve downstream reasoning and instruction-following tasks.
  arXiv  Detail & Related papers  (2026-06-26T00:16:40Z)
• Towards Closing the Autoregressive Gap in Language Modeling via Entropy-Gated Continuous Bitstream Diffusion [14.692250535148196]
  Diffusion language models (DLMs) promise parallel, order-agnostic generation but have historically lagged behind autoregressive models in sample quality and diversity.<n>Recent continuous flow and diffusion approaches over token embeddings have narrowed this gap, suggesting continuous state spaces are highly effective for language.<n>In this work, we further close the autoregressive gap by modeling text as a continuous diffusion process over fixed-width binary bitstreams.
  arXiv  Detail & Related papers  (2026-05-07T22:50:51Z)
• One-step Language Modeling via Continuous Denoising [36.18484491074519]
  We show that language models leveraging flow-based continuous denoising can outperform discrete diffusion in both quality and speed.<n>Our work calls into question the widely held hypothesis that discrete diffusion processes are necessary for generative modeling over discrete modalities.
  arXiv  Detail & Related papers  (2026-02-18T19:23:07Z)
• Self-Rewarding Sequential Monte Carlo for Masked Diffusion Language Models [58.946955321428845]
  This work presents self-rewarding sequential Monte Carlo (SMC)<n>Our algorithm stems from the observation that most existing MDLMs rely on a confidence-based sampling strategy.<n>We introduce the trajectory-level confidence as a self-rewarding signal for assigning particle importance weights.
  arXiv  Detail & Related papers  (2026-02-02T09:21:45Z)
• CDLM: Consistency Diffusion Language Models For Faster Sampling [54.886467592798]
  Diffusion Language Models (DLMs) offer a promising parallel generation paradigm but suffer from slow inference.<n>We introduce CDLM, a training-based acceleration method that simultaneously tackles both bottlenecks.<n>Experiments show CDLM achieves 3.6x-14.5x lower latency while maintaining competitive accuracy on math and coding tasks.
  arXiv  Detail & Related papers  (2025-11-24T16:21:25Z)
• Continuous Autoregressive Language Models [56.49239051750678]
  We introduce Continuous Autoregressive Language Models (CALM)<n>CALM uses a high-fidelity autoencoder to compress a chunk of K tokens into a single continuous vector.<n>We develop a comprehensive likelihood-free framework that enables robust training, evaluation, and controllable sampling.
  arXiv  Detail & Related papers  (2025-10-31T17:58:11Z)
• dParallel: Learnable Parallel Decoding for dLLMs [77.24184219948337]
  Diffusion large language models (dLLMs) offer parallel token prediction and lower inference latency.<n>Existing open-source models still require nearly token-length decoding steps to ensure performance.<n>We introduce dParallel, a simple and effective method that unlocks the inherent parallelism of dLLMs for fast sampling.
  arXiv  Detail & Related papers  (2025-09-30T16:32:52Z)
• Sequential Diffusion Language Models [110.06562906987052]
  Diffusion language models (DLMs) have strong theoretical efficiency but are limited by fixed-length decoding and incompatibility with key-value caches.<n>We introduce Next Sequence Prediction (NSP), which unifies next-token and next-block prediction.<n>We propose Sequential Diffusion Language Model (SDLM), which can retrofit pre-trained autoregressive language models (ALMs) at minimal cost.
  arXiv  Detail & Related papers  (2025-09-28T17:59:15Z)
• Accelerating Diffusion LLMs via Adaptive Parallel Decoding [50.9948753314669]
  We introduce adaptive parallel decoding (APD), a novel method that dynamically adjusts the number of tokens sampled in parallel.<n>APD provides markedly higher throughput with minimal quality degradations on downstream benchmarks.
  arXiv  Detail & Related papers  (2025-05-31T06:10:10Z)
• State Fourier Diffusion Language Model (SFDLM): A Scalable, Novel Iterative Approach to Language Modeling [0.0]
  This paper introduces a fully diffusion driven discrete text generation model built without any transformer or large convolution modules.<n>By composing local state space updates with global Fourier based mixing, the approach effectively captures both short and long range dependencies.
  arXiv  Detail & Related papers  (2025-03-16T02:17:40Z)
• Self-Refining Diffusion Samplers: Enabling Parallelization via Parareal Iterations [53.180374639531145]
  Self-Refining Diffusion Samplers (SRDS) retain sample quality and can improve latency at the cost of additional parallel compute.<n>We take inspiration from the Parareal algorithm, a popular numerical method for parallel-in-time integration of differential equations.
  arXiv  Detail & Related papers  (2024-12-11T11:08:09Z)
• Diffusion Forcing: Next-token Prediction Meets Full-Sequence Diffusion [61.03681839276652]
  Diffusion Forcing is a new training paradigm where a diffusion model is trained to denoise a set of tokens with independent per-token noise levels.<n>We apply Diffusion Forcing to sequence generative modeling by training a causal next-token prediction model to generate one or several future tokens.
  arXiv  Detail & Related papers  (2024-07-01T15:43:25Z)
• Accelerating Large Language Model Decoding with Speculative Sampling [9.851546623666588]
  speculative sampling is an algorithm for accelerating transformer decoding by enabling the generation of multiple tokens from each transformer call. We benchmark speculative sampling with Chinchilla, a 70 billion parameter language model, achieving a 2-2.5x decoding speedup in a distributed setup.
  arXiv  Detail & Related papers  (2023-02-02T18:44:11Z)
• ProDiff: Progressive Fast Diffusion Model For High-Quality Text-to-Speech [63.780196620966905]
  We propose ProDiff, on progressive fast diffusion model for high-quality text-to-speech. ProDiff parameterizes the denoising model by directly predicting clean data to avoid distinct quality degradation in accelerating sampling. Our evaluation demonstrates that ProDiff needs only 2 iterations to synthesize high-fidelity mel-spectrograms. ProDiff enables a sampling speed of 24x faster than real-time on a single NVIDIA 2080Ti GPU.
  arXiv  Detail & Related papers  (2022-07-13T17:45:43Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.