Large Language Models (LLMs), constrained by their auto-regressive nature, suffer from slow decoding. Speculative decoding methods have emerged as a promising solution to accelerate LLM decoding, attracting attention from both systems and AI research communities. Recently, the pursuit of better draft quality has driven a trend toward parametrically larger draft models, which inevitably introduces substantial computational overhead. While existing work attempts to balance the trade-off between prediction accuracy and compute latency, we address this fundamental dilemma through architectural innovation. We propose PRISM, which disaggregates the computation of each predictive step across different parameter sets, refactoring the computational pathways of draft models to successfully decouple model capacity from inference cost. Through extensive experiments, we demonstrate that PRISM outperforms all existing draft architectures, achieving exceptional acceptance lengths while maintaining minimal draft latency for superior end-to-end speedup. We also re-examine scaling laws with PRISM, revealing that PRISM scales more effectively with expanding data volumes than other draft architectures. Through rigorous and fair comparison, we show that PRISM boosts the decoding throughput of an already highly optimized inference engine by more than 2.6x.

翻译：大型语言模型（LLMs）受限于其自回归特性，存在解码速度缓慢的问题。推测解码方法作为一种加速LLM解码的有效方案，已引起系统和人工智能研究领域的广泛关注。近期，为提升草稿质量，研究趋势倾向于采用参数规模更大的草稿模型，但这不可避免地带来了显著的计算开销。现有工作试图在预测精度与计算延迟之间寻求平衡，而本研究通过架构创新从根本上解决这一矛盾。我们提出PRISM方法，其将每个预测步骤的计算分解至不同的参数集，通过重构草稿模型的计算路径，成功实现了模型容量与推理成本的解耦。大量实验表明，PRISM在保持极低草稿延迟的同时，实现了卓越的接受长度，其性能超越所有现有草稿架构，获得了优异的端到端加速效果。我们进一步基于PRISM重新审视缩放定律，发现相比其他草稿架构，PRISM在数据规模扩展时表现出更优的缩放效率。通过严格公平的对比实验，我们证明PRISM可将已高度优化的推理引擎的解码吞吐量提升2.6倍以上。