测试时强到弱能力迁移：通过脚手架实现

AI 摘要

本文研究了一种名为“强到弱脚手架”的测试时能力迁移方法，即由更强的构建模型为较弱的目标模型构建推理时脚手架，在不更新参数的情况下提升其任务表现。在四个心智理论基准测试中，该方法将目标模型的平均性能从0.49提升至0.91，主要归功于将不稳定的模型推理卸载到确定性代码、基准特定路由和严格答案格式执行。研究还发现，构建模型的推理努力与脚手架质量单调相关，且较弱的目标模型获益最大。

正文节选

AI4AI at Test-Time: Strong-to-Weak Capability Transfer via Harnesses Abstract Stronger models can build inference-time harnesses that substantially improve weaker models' task performance without parameter updates by offloading reasoning into structured code and routing. 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,

发布时间：—

抓取时间：2026-08-13 10:57

来源机构：Hugging Face