自适应并行蒙特卡洛树搜索用于高效测试时间计算扩展
Adaptive Parallel Monte Carlo Tree Search for Efficient Test-time Compute Scaling
AI总结：
本文提出负早退和自适应提升机制，通过优化MCTS执行效率，降低端到端延迟并提升吞吐量，同时保持推理准确性。
摘要：
蒙特卡洛树搜索（MCTS）是一种有效的测试时间计算扩展（TTCS）方法，用于提升大语言模型的推理性能，但其高度变化的执行时间导致实际应用中的严重长尾延迟。现有优化如正早退在有利情况下减少延迟，但当搜索无实质性进展时效果有限。我们引入负早退，修剪无生产力的MCTS轨迹，并引入自适应提升机制，重新分配回收的计算资源以减少并发搜索间的资源竞争。集成到vLLM中，这些技术显著降低了p99端到端延迟，同时提高了吞吐量并保持推理准确性。
英文摘要：
Monte Carlo Tree Search (MCTS) is an effective test-time compute scaling (TTCS) method for improving the reasoning performance of large language models, but its highly variable execution time leads to severe long-tail latency in practice. Existing optimizations such as positive early exit, reduce latency in favorable cases but are less effective when search continues without meaningful progress. We introduce {\it negative early exit}, which prunes unproductive MCTS trajectories, and an {\it adaptive boosting mechanism} that reallocates reclaimed computation to reduce resource contention among concurrent searches. Integrated into vLLM, these techniques substantially reduce p99 end-to-end latency while improving throughput and maintaining reasoning accuracy.
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。
