测试时AI4AI：通过驾驭机制实现从强到弱的能力迁移

摘要

最近关于蒸馏的研究，常常通过教师强制、在线策略蒸馏及相关训练时方法，将大模型的能力迁移到较小模型中，通常是通过更新后者的参数来实现的。在本文中，我们提出一个问题：这种迁移是否可以在测试时实现？我们研究了强到弱脚手架（strong-to-weak scaffolding）：一个更强的构建者模型能否构建推理时的辅助框架，帮助更弱的目标模型更可靠地解决问题，而无需任何参数更新？我们使用四个具有代表性的心理理论（Theory-of-Mind）基准，每个构建者模型使用5%的数据作为验证集，通过多轮迭代优化其辅助框架，最终确定的框架再在完整测试集上进行评估。实验表明，这种测试时能力迁移形式非常有效，将目标模型的平均性能从0.49提高到0.91，近乎翻倍。我们的分析显示，性能提升主要源于将不稳定的模型推理卸载到确定性代码、针对特定基准的路由策略以及严格的答案格式约束上，而非通过鼓励目标模型进行更深入的推理或更广泛的采样。我们进一步发现，构建者模型的推理投入会单调地提升辅助框架的质量；相对于构建者模型自身的能力，平台效应较为有限；而更弱的目标模型获得的收益最大。这些结果表明，推理时辅助框架的设计是对传统训练时蒸馏的重要补充，使强模型能够在无需重新训练的情况下，将认知结构迁移给更弱的模型。

English

Recent work on distillation transfers the capabilities of large models to smaller ones often by updating the latter's parameters, through teacher forcing, on-policy distillation, and related training-time methods. In this paper, we ask whether such transfer can instead occur at test time. We study strong-to-weak scaffolding: whether a stronger builder model can construct inference-time harnesses that help a weaker target model solve tasks more reliably without any parameter updates. Using four representative Theory-of-Mind benchmarks, each builder model uses 5% of the data as a validation set to iteratively refine its harness over multiple rounds, after which the finalized harness is evaluated on the full test set. Empirically, this form of test-time capability transfer is highly effective, nearly doubling average target-model performance from 0.49 to 0.91. Our analysis shows that the gains come primarily from offloading unstable model reasoning into deterministic code, benchmark-specific routing, and strict answer-format enforcement, rather than from encouraging the target model to reason more extensively or sample more broadly. We further find that builder-model reasoning effort improves harness quality monotonically, platform effects are modest relative to the builder model's own capability, and weaker target models receive the largest gains. These results suggest that inference-time harness design is an important complement to conventional training-time distillation, enabling strong models to transfer cognitive structure to weaker models without retraining.