对比大模型并行与串行采样，发现探索不足是并行更优的关键原因。
Understanding Performance Gap Between Parallel and Sequential Sampling in Large Reasoning Models
- 通过对比并行与串行采样策略，分析性能差异来源。
- 实验验证并行采样在数学和编码任务上表现更优，尤其在多轮推理中。
- 核心发现：串行采样因依赖前序结果而减少探索，导致性能下降。
大型推理模型（LRMs）在数学和编程等复杂问题上表现出色，但高质量解通常需多次采样。理论上，有两种采样策略可组合成复杂流程：串行采样与并行采样。本文首次严谨比较二者，发现尽管串行采样应具备更强表示能力，但并行采样仍表现更优。为理解其背后原因，提出三个假设：(i) 并行采样优势源于聚合操作；(ii) 串行采样受长上下文影响；(iii) 串行采样因依赖前序答案而探索不足。在Qwen3、DeepSeek-R1蒸馏模型及Gemini 2.5等多个模型家族与规模，以及数学与编程任务上的实证表明，聚合机制与上下文长度并非主因，而探索不足则显著影响性能，是导致差距的主要因素。
原文摘要 · Abstract (English)
Large Reasoning Models (LRMs) have shown remarkable performance on challenging questions, such as math and coding. However, to obtain a high quality solution, one may need to sample more than once. In principal, there are two sampling strategies that can be composed to form more complex processes: sequential sampling and parallel sampling. In this paper, we first compare these two approaches with rigor, and observe, aligned with previous works, that parallel sampling seems to outperform sequential sampling even though the latter should have more representation power. To understand the underline reasons, we make three hypothesis on the reason behind this behavior: (i) parallel sampling outperforms due to the aggregator operator; (ii) sequential sampling is harmed by needing to use longer contexts; (iii) sequential sampling leads to less exploration due to conditioning on previous answers. The empirical evidence on various model families and sizes (Qwen3, DeepSeek-R1 distilled models, Gemini 2.5) and question domains (math and coding) suggests that the aggregation and context length do not seem to be the main culprit behind the performance gap. In contrast, the lack of exploration seems to play a considerably larger role, and we argue that this is one main cause for the performance gap.
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。