CoBa: Cost-Effective Test-Time Scaling via Compute-Balanced Routing· CoBa: 计算平衡路由的高效测试时扩展
Test-time scaling is often implemented by spending more compute along one axis: sampling more solutions, extending a chain of thought, or applying a stronger evaluator. Under a fixed inference budget, these choices compete. This paper formulates test-time reasoning as a compute-allocation problem in which a system must decide whether the next unit of compute should be spent on generation, verification, or stopping. We introduce CoBa, a compute-balanced routing policy that first obtains a small set of candidates, applies cheap verification broadly, and routes uncertain or high-value candidates to stronger verification. On 3,129 example-generator evaluations spanning MATH-500, AIME 2024/2025, AMC 2023, and procedural symbolic reasoning, CoBa-Routed-Strong reaches 85.13% macro accuracy, stati
通过计算平衡路由策略提高测试时扩展效率。
- 核心方法
- 引入CoBa计算平衡路由策略，先生成少量候选答案，再广泛进行廉价验证，最后将不确定或高价值的候选答案路由到更强的验证模型。
- 适合谁读
- 研究者、工程师
- 要解决的问题
- 解决在固定推理预算下，测试时扩展中生成、验证和停止之间的计算分配问题。
- 关键实验
- 实验涉及3,129个示例生成器评估，涵盖了MATH-500、AIME 2024/2025、AMC 2023和程序符号推理任务。
- 主要贡献
- CoBa-Routed-Strong在多个数学问题数据集上达到了85.13%的宏观准确率，提高了测试时推理效率。
- 意义与局限
- 该方法在保持高准确率的同时显著降低了计算成本，但可能在更复杂的任务上效果有限。