改进大模型推理速度，让预测更准更稳。
DBLAST: Dependent Block Drafting for Stochastic Speculative Decoding
- 用低秩潜在混合建模词元依赖关系，突破独立假设限制。
- 在高熵采样下，平均通过的词元数提升30%以上。
- 适合需要高随机性推理的场景，如创意写作与复杂问答。
推测解码通过轻量级草稿器预估多个未来词元，并由目标模型验证。现有块式与扩散式草稿器通常针对贪心解码优化，假设草稿块内词元条件独立，这在非贪心推测解码中失效，因目标分布刻意引入随机性，导致多路径均可能成立。本文研究块扩散草稿器在此场景下的失配问题，发现随着目标采样分布熵增加，被接受的草稿长度显著下降。为此提出依赖块草稿器（DBLAST），基于词元位置的低秩潜在混合建模依赖关系，并采用以期望验证长度为目标的训练目标。在Qwen3-4B与Qwen3-8B上，于GSM8K、MT-Bench、HumanEval及创意写作基准测试中，本方法在高熵解码条件下持续优于独立块采样，显著提升有效生成长度。
原文摘要 · Abstract (English)
Speculative decoding accelerates large language models' inference by using a lightweight drafter to propose multiple future tokens and a target model to verify them. While recent block and diffusion-style drafters can predict several positions in a single pass, their training and sampling procedures are typically optimized for greedy decoding or assume that positions in the draft block are conditionally independent. This assumption becomes brittle in non-greedy speculative decoding, where the target distribution is deliberately stochastic and multiple continuations become plausible. We study this mismatch for block diffusion drafters and show that the accepted draft length degrades as the entropy of the target sampling distribution increases. We propose a dependent block drafter based on a low-rank latent mixture over token positions, complemented by an acceptance-oriented training objective that directly targets the expected verified length. Experiments with Qwen3-4B and Qwen3-8B on GSM8K, MT-Bench, HumanEval, and creative-writing benchmarks show that our approach, namely DBLast, consistently improves accepted length over independent block sampling, especially in higher-entropy decoding regimes.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。