发表机构
NAVER AI Lab; Korea University; NAVER AI Search Platform（NAVER人工智能实验室; 高丽大学; NAVER人工智能搜索平台）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出感知验证训练（VAT）框架，将其应用于EAGLE-3、DFlash等方法，在多款大语言模型上使推测解码的平均接受长度提升最多11.4%，推理速度提升最多8.7%。
AI 中文摘要
推测解码通过使用草稿模型生成候选token，并由目标模型在单次前向传播中进行验证，从而加速大语言模型推理。验证过程按顺序进行，且会丢弃从第一个被拒绝的位置开始的所有位置，然而现有的草稿模型训练依赖于对目标模型的token级模仿，采用固定的每位置权重，该权重未反映上述两种特性。我们提出感知验证训练（VAT），这是一种即插即用框架，它在每个训练步骤中模拟验证过程，并将产生的接受和拒绝模式转化为监督信号。VAT包含两个组件：（i）验证头，一个轻量级的联合训练二分类器，用于在每个位置是否通过顺序验证方面监督草稿模型；（ii）验证自适应权重，它替换了固定的权重调度，方法是在每个样本的第一个拒绝点之前保持全权重，并从该点重新开始衰减。VAT仅修改训练目标，因此可以叠加在现有方法之上，而无需改变草稿架构、目标模型或推理过程。将VAT应用于EAGLE-3和DFlash模型，在Qwen3-4B、Qwen3-8B和LLaMA-3.1-8B上进行测试，结果显示平均接受长度最多提升11.4%， wall-clock加速比最多提升8.7%，在数学、代码和聊天基准测试中均取得一致的性能提升。代码将在此处提供：this https URL
英文摘要
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
Comments14 pages
