Speculative Pipeline Decoding: Higher-Accuracy Drafting with Hidden Latency via Pipeline Parallelism

תקציר מקורי באנגליתarXiv:2605.30852v3 Announce Type: replace Abstract: Speculative Decoding (SD) accelerates low-concurrency LLM inference with a draft-then-verify paradigm. Mainstream methods, however, rely on multi-token prediction, which incurs compounding prediction difficulty and exposed draft latency. We propose Speculative Pipeline Decoding (SPD), which partitions the target LLM into $n$ pipeline stages so that $n$ tokens of a single sequence advance in parallel. To keep the pipeline saturated, a Pipeline Draft Module (PDM) aggregates multi-depth target features to predict the next token and runs concurrently with each pipeline step, yielding bounded prediction difficulty, higher acceptance, and hidden draft latency. Experiments show that SPD achieves higher theoretical and wall-clock speedup than EAG