研究人员开发了一个新框架，以提高大型推理模型（LRMs）在复杂数学任务上的性能。这种无需训练的方法利用输出不一致性作为信号，为每个实例动态选择最合适的测试时扩展策略。该系统将一致的案例路由到轻量级解析，将中度不一致路由到多数投票，并将高度模糊的问题路由到基于重写的重构。实验表明，与现有技术相比，该方法在提高准确性的同时，将计算成本降低了 3-7%。 AI
影响 通过动态调整测试时策略，提高了大型语言模型在数学任务上的推理准确性和效率。
排序理由 关于改进大型语言模型推理的新方法的学术论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2604.26644v1 Announce Type: new Abstract: Large Reasoning Models (LRMs) achieve strong performance on mathematical reasoning tasks but remain unreliable on challenging instances. Existing test-time scaling methods, such as repeated sampling, self-correction, and tree search…
Large Reasoning Models (LRMs) achieve strong performance on mathematical reasoning tasks but remain unreliable on challenging instances. Existing test-time scaling methods, such as repeated sampling, self-correction, and tree search, improve performance at the cost of increased c…