分享更多，搜索更少：面向高效测试时间扩展的协作并行思考
Share More, Search Less: Collaborative Parallel Thinking for Efficient Test-Time Scaling
AI总结：
提出一种无需训练的协作并行思考框架，通过在并行分支间共享搜索信息来减少冗余探索，从而在测试时间扩展中实现更优的准确率-延迟帕累托边界。
摘要：
测试时间扩展（TTS）通过分配额外的推理计算来探索解空间，从而增强大型语言模型的推理能力。然而，现有的并行TTS方法通常在搜索过程中保持分支隔离：中间发现保持分支私有，无法及时指导其他分支。这种信息隔离导致大量冗余探索，因为分支反复重新发现其他地方已有的信息，并且需要更多搜索步骤来收集做出正确回答所需的完整决策信息。为弥补这一差距，我们提出协作并行思考（CPT），一种无需训练的推理框架，能够在并行分支间实现搜索时信息共享。CPT从正在运行的分支中提取紧凑的中间信息，维护一个去重的查询级信息池，并通过输入上下文广播池条目，使得后续搜索步骤中的每个分支能够重用其他分支的发现，而不是重新发现相同信息。实验上，在HMMT和AIME基准测试上的结果表明，CPT在不同rollout预算和模型规模下，相比强基线建立了更强的准确率-延迟帕累托前沿，突显了搜索时协作作为高效并行TTS的一个有效方向。
英文摘要：
Test-Time Scaling (TTS) enhances the reasoning capabilities of large language models by allocating additional inference compute to explore the solution space. However, existing parallel TTS methods typically keep branches isolated during search: intermediate discoveries remain branch-private and cannot guide other branches in time. This information isolation causes substantial redundant exploration, as branches repeatedly rediscover information already found elsewhere and require more search steps to collect complete decision information needed to reach correct answers. To bridge this gap, we propose \textbf{Collaborative Parallel Thinking (CPT)}, a training-free inference framework that enables search-time information sharing across parallel branches. CPT extracts compact intermediate information from ongoing branches, maintains a deduplicated query-level information pool, and broadcasts pool entries through the input context, allowing each branch in subsequent search steps to reuse discoveries made by other branches rather than rediscover the same information. Empirically, experiments on HMMT and AIME benchmarks show that CPT establishes a stronger accuracy--latency Pareto frontier than strong baselines across rollout budgets and model scales, highlighting search-time collaboration as an effective direction for efficient parallel TTS.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
