Verification-Aware Training for Speculative Decoding

• URL: http://arxiv.org/abs/2608.30135v1
• Date: Mon, 31 Aug 2026 01:42:10 GMT
• Title: Verification-Aware Training for Speculative Decoding
• Abstract summary: 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.
• Score: 57.84976863792784
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Speculative decoding accelerates large language model inference by using a draft model to generate candidate tokens, which are verified by the target model in a single forward pass. Verification proceeds sequentially and discards every position from the first rejection onward, yet existing draft training relies on token-level imitation of the target with a fixed per-position weighting that reflects neither property. 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. VAT consists of two components: (i) a verification head, a lightweight jointly trained binary classifier that supervises the draft model on whether each position survives sequential verification; (ii) verification-adaptive weighting, which replaces the fixed weighting schedule by keeping full weight up to each sample's first rejection point and re-anchoring the decay to start there. 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. Applied to EAGLE-3 and DFlash on Qwen3-4B, Qwen3-8B, and LLaMA-3.1-8B, 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. Code will be available at https://github.com/naver-ai/vat

Related papers

• ReTrace: Rejected-Trajectory Conditioning for Speculative Decoding [46.17292752074153]
  We show that rejected positions in a rejected suffix may still align with the target continuation.<n>textbfReTrace, a rejected-trajectory conditioning method, conditions each draft block on the rejected suffix from the previous round.<n>ReTrace consistently improves average acceptance length and end-to-end decoding speed over its DFlash backbone.
  arXiv  Detail & Related papers  (2026-08-30T12:20:54Z)
• Parallel Prefix Verification for Speculative Generation [10.689879928713564]
  PARSE (PArallel pRefix Speculative Engine) is a speculative generation framework that accelerates large language model (LLM) inference.<n>We introduce PARSE, a speculative generation framework that accelerates large language model (LLM) inference by parallelizing prefix verification on a semantic level.
  arXiv  Detail & Related papers  (2026-05-05T19:56:37Z)
• From Tokens to Steps: Verification-Aware Speculative Decoding for Efficient Multi-Step Reasoning [18.37034672013596]
  SpecGuard is a speculative decoding framework that performs step-level verification using only model-internal signals.<n> Experiments show that SpecGuard improves accuracy by 3.6% while reducing latency by 11%, outperforming both SD and reward-guided SD.
  arXiv  Detail & Related papers  (2026-04-16T17:20:13Z)
• Training-free Dropout Sampling for Semantic Token Acceptance in Speculative Decoding [13.249778063956917]
  Speculative decoding accelerates large language model inference by proposing tokens with a lightweight draft model.<n>This work introduces DropMatch, a novel approach that matches draft tokens to the predictive distribution of the target model.<n> Experiments across multiple benchmarks demonstrate that our approach increases acceptance length while maintaining competitive task performance.
  arXiv  Detail & Related papers  (2026-02-11T04:53:33Z)
• Training-Free Loosely Speculative Decoding: Accepting Semantically Correct Drafts Beyond Exact Match [21.810129153556044]
  Training-Free Loosely Speculative Decoding (FLy) is a novel method that loosens the rigid verification criterion.<n>We show that FLy preserves more than 99% of the target model's accuracy while achieving an average 2.81x speedup.
  arXiv  Detail & Related papers  (2025-11-28T08:23:30Z)
• DiffuSpec: Unlocking Diffusion Language Models for Speculative Decoding [66.40658898418316]
  We present DiffuSpec, a training-free drop-in framework that uses a pretrained diffusion language model (DLM) to produce multi-token drafts in a single forward pass.<n>Across benchmarks, DiffuSpec yields up to 3x wall-clock speedup, establishing diffusion-based drafting as a robust alternative to autoregressive drafters for speculative decoding.
  arXiv  Detail & Related papers  (2025-09-28T07:00:15Z)
• Rethinking Classifier Re-Training in Long-Tailed Recognition: A Simple Logits Retargeting Approach [102.0769560460338]
  We develop a simple logits approach (LORT) without the requirement of prior knowledge of the number of samples per class. Our method achieves state-of-the-art performance on various imbalanced datasets, including CIFAR100-LT, ImageNet-LT, and iNaturalist 2018.
  arXiv  Detail & Related papers  (2024-03-01T03:27:08Z)
• Q-REG: End-to-End Trainable Point Cloud Registration with Surface Curvature [81.25511385257344]
  We present a novel solution, Q-REG, which utilizes rich geometric information to estimate the rigid pose from a single correspondence. Q-REG allows to formalize the robust estimation as an exhaustive search, hence enabling end-to-end training. We demonstrate in the experiments that Q-REG is agnostic to the correspondence matching method and provides consistent improvement both when used only in inference and in end-to-end training.
  arXiv  Detail & Related papers  (2023-09-27T20:58:53Z)
• Certifiable 3D Object Pose Estimation: Foundations, Learning Models, and Self-Training [23.802602957611676]
  We consider a certifiable object pose estimation problem, where -- given a partial point cloud of an object -- the goal is to provide a certificate of correctness for the resulting estimate. We propose C-3PO, a semantic-keypoint-based pose estimation model, augmented with the two certificates.
  arXiv  Detail & Related papers  (2022-06-22T17:06:39Z)
• Aligning Pretraining for Detection via Object-Level Contrastive Learning [57.845286545603415]
  Image-level contrastive representation learning has proven to be highly effective as a generic model for transfer learning. We argue that this could be sub-optimal and thus advocate a design principle which encourages alignment between the self-supervised pretext task and the downstream task. Our method, called Selective Object COntrastive learning (SoCo), achieves state-of-the-art results for transfer performance on COCO detection.
  arXiv  Detail & Related papers  (2021-06-04T17:59:52Z)
• Pre-training Is (Almost) All You Need: An Application to Commonsense Reasoning [61.32992639292889]
  Fine-tuning of pre-trained transformer models has become the standard approach for solving common NLP tasks. We introduce a new scoring method that casts a plausibility ranking task in a full-text format. We show that our method provides a much more stable training phase across random restarts.
  arXiv  Detail & Related papers  (2020-04-29T10:54:40Z)

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.