并行推理后平均得分，提升代码与深度研究任务表现
Think in Parallel, Answer as One: Logit Averaging for Open-Ended Reasoning
- 运行多个并行推理路径，同步点平均下一词概率
- 代码生成任务中通过率提升超7%，最高+8.28%
- 无需训练，兼容主流解码策略，适合长文本生成
多数投票在封闭式问答中表现优异，但难以应用于开放式推理（如代码生成、网络深度调研），因完整答案的“多数”无定义。本文提出ThinkMerge，一种无需训练、即插即用的解码策略：并行运行K条推理路径，在同步点平均其下一词概率，生成单一连贯输出。该方法可无缝集成vLLM/SGLang，兼容Top-p/Top-k等标准解码方式。实验表明，其在AIME和GPQA上表现媲美甚至超越多数投票；在开放式代码任务上持续提升：LiveCodeBench（难）中，DeepCoder-14B-Preview pass@1提升+8.28%，Qwen3-8B提升+7.58%。此外，该方法还显著改进基于网络的深度调研代理（如WebSailor-7B/32B），在GAIA、BrowseComp-en/zh和XbenchDeepSearch上均取得提升。结果证明，无需对完整输出进行投票，并行测试时缩放仍可有效提升开放式推理性能。
原文摘要 · Abstract (English)
Majority voting has proven effective for close-ended question answering by aggregating parallel reasoning traces. However, it is not directly applicable to open-ended reasoning, such as code generation and web-based deep research, where a "majority" over complete solutions is ill-defined. We introduce ThinkMerge, a training-free, plug-and-play decoding strategy that runs K parallel reasoning traces and averages their next-token logits at synchronization points to produce a single coherent output. ThinkMerge integrates seamlessly with vLLM/SGLang and remains compatible with standard decoding techniques such as Top-p/Top-k. Empirically, it matches or surpasses majority voting on AIME and GPQA, while delivering consistent gains on open-ended coding tasks: on LiveCodeBench (hard), pass@1 improves by +8.28% for DeepCoder-14B-Preview and +7.58% for Qwen3-8B. Beyond code, we further show that ThinkMerge improves web-based deep-research agents (e.g., WebSailor-7B/32B) across GAIA, BrowseComp-en/zh, and XbenchDeepSearch. These results demonstrate that parallel test-time scaling can benefit open-ended reasoning without relying on voting over complete outputs.
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。