VVS：通过部分验证跳过加速视觉自回归生成的投机解码
VVS: Accelerating Speculative Decoding for Visual Autoregressive Generation via Partial Verification Skipping
把 LLM 里的投机解码搬到视觉自回归生成上，并首次提出「跳过验证步」：用动态截断的免验证词元选择器 + 词元级特征缓存复用 + 细粒度跳步调度，直接减少目标模型前向次数（2.8×），对做视觉 AR 生成（如 VAR 类模型）落地加速的人是一套可直接复用的推理框架。
原文依据 we explore verification skipping in the SD process for the first time to explicitly cut the number of target model forward passes
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论视觉自回归（AR）生成模型在图像生成上展现出强大潜力，但其下一词元预测范式带来了可观的推理延迟。尽管投机解码（SD）已被证明能有效加速视觉 AR 模型，其「草稿一步、验证一步」的范式无法直接减少前向次数，从而限制了加速潜力。受视觉词元可互换性的启发，我们首次在 SD 过程中探索验证跳过，以显式削减目标模型的前向次数，进而降低推理延迟。通过分析草稿阶段的特性，我们观察到验证冗余与陈旧特征可复用性是在提升速度的同时对免验证步骤保持生成质量的关键因素。受这两点观察启发，我们提出一种新的 SD 框架 VVS，通过部分验证跳过加速视觉 AR 模型，它集成了三个互补模块：（1）带动态截断的免验证词元选择器，（2）词元级特征缓存与复用，（3）细粒度的跳步调度。由此，VVS 在保持有竞争力的生成质量的同时，将目标模型前向次数相对原始 AR 解码减少 2.8 倍，相比传统 SD 框架提供了更优的速度—质量权衡，并展现出重塑 SD 范式的强大潜力。我们的代码见 https://github.com/HyattDD/VVS。
英文原文摘要 1620 字符
Visual autoregressive (AR) generation models have demonstrated strong potential for image generation, yet their next-token-prediction paradigm introduces considerable inference latency. Although speculative decoding (SD) has been proven effective for accelerating visual AR models, its "draft one step, then verify one step" paradigm prevents a direct reduction in the number of forward passes, limiting its acceleration potential. Motivated by the interchangeability of visual tokens, we explore verification skipping in the SD process for the first time to explicitly cut the number of target model forward passes, thereby reducing inference latency. By analyzing the characteristics of the drafting stage, we observe that verification redundancy and stale feature reusability are key factors to maintain generation quality while improving speed for verification-free steps. Inspired by these two observations, we propose a novel SD framework VVS to accelerate \underline \text v isual AR model via partial \underline \text v erification \underline \text s kipping, which integrates three complementary modules: (1) a verification-free token selector with dynamically truncation, (2) token level feature caching and reuse, and (3) fine-grained skipped step scheduling. Consequently, VVS reduces the number of target model forward passes by 2.8x relative to vanilla AR decoding while maintaining competitive generation quality, offering a superior speed-quality trade-off over conventional SD frameworks and revealing strong potential to reshape the SD paradigm. Our code is available at https://github.com/HyattDD/VVS.