Beyond Scattered Acceptance: Fast and Coherent Inference for DLMs via Longest Stable Prefixes

• URL: http://arxiv.org/abs/2603.05454v1
• Date: Thu, 05 Mar 2026 18:25:26 GMT
• Title: Beyond Scattered Acceptance: Fast and Coherent Inference for DLMs via Longest Stable Prefixes
• Authors: Pengxiang Li, Joey Tsai, Hongwei Xue, Kunyu Shi, Shilin Yan,
• Abstract summary: Longestahead Prefix (LSP) scheduler is a training-free and model-agnostic inference paradigm based on monolithic prefix absorption.<n>LSP evaluates token stability via a single forward pass, dynamically identifies a contiguous left-aligned block of stable predictions.<n>It snaps its boundary to natural linguistic or structural acceptances before an atomic commitment.
• Score: 10.877713536966601
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Diffusion Language Models (DLMs) promise highly parallel text generation, yet their practical inference speed is often bottlenecked by suboptimal decoding schedulers. Standard approaches rely on 'scattered acceptance'-committing high confidence tokens at disjoint positions throughout the sequence. This approach inadvertently fractures the Key-Value (KV) cache, destroys memory locality, and forces the model into costly, repeated repairs across unstable token boundaries. To resolve this, we present the Longest Stable Prefix (LSP) scheduler, a training-free and model-agnostic inference paradigm based on monolithic prefix absorption. In each denoising step, LSP evaluates token stability via a single forward pass, dynamically identifies a contiguous left-aligned block of stable predictions, and snaps its boundary to natural linguistic or structural delimiters before an atomic commitment. This prefix-first topology yields dual benefits: systemically, it converts fragmented KV cache updates into efficient, contiguous appends; algorithmically, it preserves bidirectional lookahead over a geometrically shrinking active suffix, drastically reducing token flip rates and denoiser calls. Extensive evaluations on LLaDA-8B and Dream-7B demonstrate that LSP accelerates inference by up to 3.4x across rigorous benchmarks including mathematical reasoning, code generation, multilingual (CJK) tasks, and creative writing while matching or slightly improving output quality. By fundamentally restructuring the commitment topology, LSP bridges the gap between the theoretical parallelism of DLMs and practical hardware efficiency.

Related papers

• Prism: Efficient Test-Time Scaling via Hierarchical Search and Self-Verification for Discrete Diffusion Language Models [96.0074341403456]
  Inference-time compute has re-emerged as a practical way to improve LLM reasoning.<n>Most test-time scaling (TTS) algorithms rely on autoregressive decoding.<n>We propose Prism, an efficient TTS framework for dLLMs.
  arXiv  Detail & Related papers  (2026-02-02T09:14:51Z)
• CD4LM: Consistency Distillation and aDaptive Decoding for Diffusion Language Models [27.070045950001532]
  CD4LM is a framework that decouples training from inference.<n>On GSM8K, CD4LM matches the LLaDA baseline with a 5.18x wall-clock speedup.
  arXiv  Detail & Related papers  (2026-01-05T16:09:22Z)
• WeDLM: Reconciling Diffusion Language Models with Standard Causal Attention for Fast Inference [44.87788417755154]
  We propose WeDLM, a diffusion decoding framework built entirely on standard causal attention.<n>We show that WeDLM preserves the quality of strong AR backbones while delivering substantial speedups.
  arXiv  Detail & Related papers  (2025-12-28T01:25:48Z)
• Beyond Surface Reasoning: Unveiling the True Long Chain-of-Thought Capacity of Diffusion Large Language Models [54.81955614221652]
  parallel decoding, which enables simultaneous token updates, conflicts with the causal order often required for rigorous reasoning.<n> Behavioral analyses in both simple and complex reasoning tasks show thatDLLMs exhibit genuine parallelism only for directly decidable outputs.<n>We propose several practical mitigations, parallel-oriented prompting, diffusion early stopping, and parallel scaling, to reduce PSC-induced ineffectiveness and inefficiencies.
  arXiv  Detail & Related papers  (2025-10-10T16:58:14Z)
• Accelerating Diffusion LLM Inference via Local Determinism Propagation [27.751279909685604]
  LocalLeap is a training-free adaptive parallel decoding strategy.<n>It achieves 6.94$times$ throughput improvements and reduces decoding steps to just 14.2% of the original requirement.
  arXiv  Detail & Related papers  (2025-10-08T14:39:34Z)
• Free Draft-and-Verification: Toward Lossless Parallel Decoding for Diffusion Large Language Models [8.407364705777587]
  We introduce Free Draft-and-Verification (FreeDave), a novel fast decoding algorithm tailored forDLLMs.<n>FreeDave is proven to boost the inference throughput up to $3.78times$ without performance degradation.
  arXiv  Detail & Related papers  (2025-09-30T21:28:04Z)
• 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)
• READER: Retrieval-Assisted Drafter for Efficient LLM Inference [0.0386965802948046]
  Autoregressive Language Models instantiate a factorized likelihood over token sequences, yet their strictly sequential decoding process imposes an intrinsic lower bound on latency inference.<n>This bottleneck has emerged as a central obstacle to the scalable deployment of large-scale generative models.<n>We present READER, a speculative decoding framework that bypasses the training of the auxiliary draft model.
  arXiv  Detail & Related papers  (2025-08-12T16:47:48Z)
• R-Stitch: Dynamic Trajectory Stitching for Efficient Reasoning [80.104336426172]
  Chain-of-thought (CoT) enhances problem-solving ability of large language models.<n>CoT incurs substantial inference cost due to long autoregressive trajectories.<n>We introduce R-Stitch, a training-free hybrid decoding framework.
  arXiv  Detail & Related papers  (2025-07-23T08:14:36Z)
• COrAL: Order-Agnostic Language Modeling for Efficient Iterative Refinement [80.18490952057125]
  Iterative refinement has emerged as an effective paradigm for enhancing the capabilities of large language models (LLMs) on complex tasks. We propose Context-Wise Order-Agnostic Language Modeling (COrAL) to overcome these challenges. Our approach models multiple token dependencies within manageable context windows, enabling the model to perform iterative refinement internally.
  arXiv  Detail & Related papers  (2024-10-12T23:56:19Z)
• Towards Continual Learning Desiderata via HSIC-Bottleneck Orthogonalization and Equiangular Embedding [55.107555305760954]
  We propose a conceptually simple yet effective method that attributes forgetting to layer-wise parameter overwriting and the resulting decision boundary distortion. Our method achieves competitive accuracy performance, even with absolute superiority of zero exemplar buffer and 1.02x the base model.
  arXiv  Detail & Related papers  (2024-01-17T09:01:29Z)
• Confident Adaptive Language Modeling [95.45272377648773]
  CALM is a framework for dynamically allocating different amounts of compute per input and generation timestep. We demonstrate the efficacy of our framework in reducing compute -- potential speedup of up to $times 3$ -- while provably maintaining high performance.
  arXiv  Detail & Related papers  (2022-07-14T17:00:19Z)

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.