Spec-AUF: Accept-Until-Fail Training under Train-Inference Misalignment for Masked Block Drafters

• URL: http://arxiv.org/abs/2607.01893v1
• Date: Thu, 02 Jul 2026 08:44:04 GMT
• Title: Spec-AUF: Accept-Until-Fail Training under Train-Inference Misalignment for Masked Block Drafters
• Abstract summary: Block (DLM-style) drafters predict the whole block in parallel, which is fast but trained with a full-block cross-entropy.<n>Recent acceptance-aware objectives patch this by reweighting the full-block loss.<n>AUF approximates that prefix-sensitive supervision on the loss side by keeping the cross-entropy support only through the drafter's first predicted failure.
• Score: 3.9513930662588863
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Speculative decoding accelerates autoregressive generation by drafting a block of tokens that the target model verifies left-to-right, committing only the longest accepted prefix. Block (DLM-style) drafters predict the whole block in parallel, which is fast but trained with a full-block cross-entropy that supervises every position against the gold continuation -- even though inference discards every token after the first rejection. Recent acceptance-aware objectives patch this by reweighting the full-block loss; we instead use teacher-forced learning as a motivation for how supervision should concentrate on the accepted prefix. A mask-only block drafter has no input-side channel for gold-prefix conditioning, so AUF approximates that prefix-sensitive supervision on the loss side by keeping the cross-entropy support only through the drafter's first predicted failure. AUF is a single, detached change to the CE support -- no auxiliary objective, no verifier rollouts, and no change to the inference pipeline or the exactness contract. Within fixed drafter backbones and serving settings on Qwen3-8B, AUF raises the DFlash drafter's average emitted length $τ$, averaged over six benchmarks, from 2.40 to 2.61, with a gain on every benchmark, and transfers to Domino's two-branch head (2.56 to 2.68). Two findings sharpen the picture: the decay-only baseline reaches higher token accuracy on the shared block mask yet decodes worse, and on DFlash, once AUF truncates the support, the standard exponential position-decay weighting becomes empirically inert.

Related papers

• Ceiling-Clipped Acceptance Histograms Indicate Stranded Speed-up in Block-Diffusion Speculative Decoding [0.008640713300591884]
  Speculative decoding speeds up generation with an efficient draft model (drafter)<n>High-acceptance block-diffusion drafters such as DFlash and DFlare fill an entire block in one parallel pass.<n>We post-train the drafter on a longer block with a short curriculum that emphasizes the newly exposed positions.
  arXiv  Detail & Related papers  (2026-08-31T08:23:14Z)
• Verification-Aware Training for Speculative Decoding [57.84976863792784]
  We introduce Verification-Aware Training (VAT), a plug-in framework that simulates verification at every training step and turns the resulting accept and reject patterns into supervision.<n>VAT modifies only the training objective, so it can be layered on top of existing methods without changing the draft architecture, the target model, or the inference procedure.<n>VAT improves average acceptance length by up to 11.4% and wall-clock speedup by up to 8.7%, with consistent gains across math, code, and chat benchmarks.
  arXiv  Detail & Related papers  (2026-08-31T01:42:10Z)
• From Positionwise Confidence to Prefix Scheduling: Verifier Skipping in Speculative Decoding [41.41611499810411]
  Speculative diffusion decoding (SDD) generates every position in a draft block in parallel with a discrete diffusion model.<n>This paper recognizes that this creates a new control handle: whether to invoke the verifier at all.
  arXiv  Detail & Related papers  (2026-08-14T18:00:08Z)
• CORA-Diff: Confidence-Oriented Residual Acceptance for Efficient Diffusion Language Model Inference [24.176666448303095]
  Diffusion language models update many tokens in parallel, yet practical decoders often use a fixed denoising horizon.<n>We propose CORA-Diff, a training-free method that preserves the original transfer rule and applies confidence-and-persistence gating only to positions that rule leaves unresolved.<n>Results show that native confidence and persistence enable reliable residual acceptance, reducing repeated denoising while preserving task quality.
  arXiv  Detail & Related papers  (2026-07-31T07:32:10Z)
• FlowBlock: Wavefront-Parallel Decoding for Self-Correcting Diffusion Language Models [16.404926372068047]
  Block-wise diffusion large language models (dLLMs) decode sequentially at the block level, enabling effective KV-cache reuse across blocks but making inter-block decoding strictly serial.<n>We propose textbfflowblock, a training-free parallel decoding framework built on two mechanisms.<n>It improves tokens per second (TPS) over LLaDA-2.1 and LLaDA-2.0, two serial block-wise dLLMs, by up to 2.95$times$ and 4.01$times$, while reducing latency by up to 53.6% and 77.1%
  arXiv  Detail & Related papers  (2026-07-20T08:05:03Z)
• Teaching Diffusion to Speculate Left-to-Right [3.205247598097648]
  Large language models (LLMs) achieve remarkable performance across a wide range of tasks, but their autoregressive decoding process incurs substantial inference costs.<n> Speculative decoding addresses this bottleneck by employing a lightweight draft model to propose multiple future tokens that are subsequently verified in parallel by a larger target model.<n>Recent work has demonstrated that diffusion language models are well suited for this setting, as they can generate entire blocks of draft tokens in parallel.
  arXiv  Detail & Related papers  (2026-06-10T01:21:56Z)
• D^2SD: Accelerating Speculative Decoding with Dual Diffusion Draft Models [27.44373450962651]
  Recent diffusion-based drafters generate an entire block of tokens in parallel but usually commit to a single draft sequence per verification.<n>We propose D2SD, a dual diffusion draft speculative decoding framework that organizes candidates into a confidence-guided prefix tree.<n>The resulting shared-guided candidates are jointly verified via cascade attention.
  arXiv  Detail & Related papers  (2026-06-03T04:48:00Z)
• OmniOPD: Logit-Free On-Policy Distillation via Speculative Verification [21.095025092772257]
  On-Policy Distillation (OPD) trains a student model on its own generative trajectories under dense token-level feedback from a stronger teacher.<n>We introduce OmniOPD, a novel framework that addresses both limitations through a logit-free, chunk-level supervision signal.
  arXiv  Detail & Related papers  (2026-05-31T22:31:15Z)
• Prefix-Adaptive Block Diffusion for Efficient Document Recognition [52.15352911463151]
  Block Diffusion Models (BDMs) support parallel generation, flexible-length output, and KV caching, making them promising for efficient document parsing.<n>We propose the Prefix- Block Diffusion Model (PA-BDM) which replaces intra-block bidirectional denoising with causal denoising from prefix to suffix.<n> Experiments show that the 3B PA-BDM achieves higher recognition scores on several benchmarks and improves inference throughput by 71.6% over the 2.5B MinerU-Diffusion.
  arXiv  Detail & Related papers  (2026-05-16T07:50:13Z)
• D-PACE: Dynamic Position-Aware Cross-Entropy for Parallel Speculative Drafting [59.204113363599994]
  Speculative decoding accelerates inference by having a small drafter propose tokens that a larger target model verifies in parallel.<n>Recent diffusion-based parallel drafters such as DFlash predict the full B-token block in one forward pass, enabling deeper drafters and longer accepted blocks.<n>We derive per-position training weights from a differentiable surrogate of expected accepted draft length, matching the weight of each position to its log-probability contribution.<n>Across six benchmarks, two Qwen3-4B draft depths, two decoding temperatures, and two additional target models, D-PACE consistently improves both wall-clock speedup and average
  arXiv  Detail & Related papers  (2026-05-12T06:27:57Z)
• From Next-Token to Next-Block: A Principled Adaptation Path for Diffusion LLMs [58.640039233470766]
  We show that principled AR-to-block-diffusion adaptation is an effective and compute-efficient alternative to training DLMs from scratch.<n> NBDiff-7B (Base and Instruct) could inherit the long-context modeling and reasoning capabilities, and achieve state-of-the-art performance.
  arXiv  Detail & Related papers  (2025-12-07T10:28:21Z)
• Pipeline Parallelism is All You Need for Optimized Early-Exit Based Self-Speculative Decoding [73.67253077506672]
  Large language models (LLMs) deliver impressive generation quality, but incur very high inference cost.<n>Early-exit based self-speculative decoding (EESD) has emerged to mitigate this cost.<n>We propose Pipeline-Parallel Self-Speculative Decoding (PPSD) that fully pipelines the draft and verification work.
  arXiv  Detail & Related papers  (2025-09-19T04:51:41Z)
• Blockwise SFT for Diffusion Language Models: Reconciling Bidirectional Attention and Autoregressive Decoding [60.06816407728172]
  Discrete diffusion language models have shown strong potential for text generation.<n>Standard supervised fine-tuning misaligns with semi-autoregressive inference.<n>We propose Blockwise SFT, which partitions responses into fixed-size blocks.
  arXiv  Detail & Related papers  (2025-08-27T02:49:33Z)

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.