Speculative Decoding for Multi-Sample Inference
This work addresses efficiency challenges in sampling-based reasoning for AI systems, representing a paradigm shift rather than an incremental improvement.
The paper tackled the problem of inefficient multi-sample inference in speculative decoding by proposing a novel method that exploits consensus across parallel generation paths, resulting in a substantial improvement in draft acceptance rates and reduced latency on mathematical reasoning benchmarks.
We propose a novel speculative decoding method tailored for multi-sample reasoning scenarios, such as self-consistency and Best-of-N sampling. Our method exploits the intrinsic consensus of parallel generation paths to synthesize high-quality draft tokens without requiring auxiliary models or external databases. By dynamically analyzing structural patterns across parallel reasoning paths through a probabilistic aggregation mechanism, it identifies consensus token sequences that align with the decoding distribution. Evaluations on mathematical reasoning benchmarks demonstrate a substantial improvement in draft acceptance rates over baselines, while reducing the latency in draft token construction. This work establishes a paradigm shift for efficient multi-sample inference, enabling seamless integration of speculative decoding with sampling-based reasoning techniques.