ParaTempo：基於時序置信度的高效並行推理

摘要

並行推理透過探索多條解題路徑來提升大型推理模型的準確度與穩健性，但其計算成本會隨推理深度與分支數量增加而增長。現有的平行路徑管理方法通常依賴最終答案共識、局部詞元置信度或孤立的中間探測。然而，這些訊號往往有所延遲、與實際推理進展關聯薄弱，或因過於嘈雜而難以進行動態的分支層級控制。為了解決這些限制，我們提出了 ParaTempo，一種免訓練的非同步並行推理框架。ParaTempo 由時間置信度驅動，這是一種衡量答案空間收斂程度的分支局部度量。每個分支會被定期探測以獲得暫定答案的機率分佈，而時間置信度則量化近期中間探測集中於主導答案的銳利程度。一旦累積足夠證據，ParaTempo 便僅依據此單一訊號驅動其整個控制流程：低置信度分支被剪枝、持續固守主導答案的分支提前退出、釋放的計算資源透過派生新分支重新分配，且當置信度加權投票集中時全局停止生成。無需在推理軌跡間進行同步，ParaTempo 即可基於分支層級收斂自適應地分配計算。在具有挑戰性的數學與科學推理基準測試上的實驗顯示，ParaTempo 在維持具競爭力的準確度的同時，將平均延遲降低了 21.8%–32.2%，總詞元使用量降低了 18.1%–30.3%。此外，時間置信度在未來分支收斂的預測能力與時間穩定性上，均優於詞元層級與瞬時訊號。

English

Parallel reasoning improves the accuracy and robustness of large reasoning models by exploring multiple solution paths, but its computational cost grows with reasoning depth and branch count. Existing methods for managing these parallel paths typically rely on final-answer consensus, local token confidence, or isolated intermediate probes. However, these signals are often delayed, weakly tied to actual reasoning progress, or too noisy for dynamic, branch-level control. To address these limitations, we introduce ParaTempo, a training-free asynchronous parallel reasoning framework. ParaTempo is driven by temporal confidence, a branch-local measure of answer-space convergence. Each branch is periodically probed for a tentative answer probability distribution, and temporal confidence quantifies how sharply the recent intermediate probes concentrate on a dominant answer. Once sufficient evidence has accumulated, ParaTempo drives its entire control process from this single signal: low-confidence branches are pruned, branches that persistently commit to their dominant answer are retired early, freed computation is reallocated by forking new branches, and generation stops globally once the confidence-weighted vote concentrates. Without requiring synchronization among reasoning trajectories, ParaTempo adaptively allocates computation based on branch-level convergence. Experiments on challenging mathematical and scientific reasoning benchmarks show that ParaTempo reduces average latency by 21.8-32.2% and total token usage by 18.1-30.3% while maintaining competitive accuracy. Moreover, temporal confidence exhibits stronger temporal stability and predictive power for future branch convergence than token-level and instantaneous signals.