用自然语言描述函数，自动生成可复用的神经小模型。
Compile by Training: Turning Natural-Language Specifications into Local Neural Functions
- 用教师模型生成任务样本，训练小型适配器实现编译。
- 在FuzzyBench-Hard上达到83.6%语义准确率，优于快速编译器。
- 适合需要低延迟、可部署、可组合的智能应用开发者。
许多常见文本函数易描述却难用规则实现，而每次调用大型远程模型会带来重复成本、延迟和依赖问题。本文提出“编译即训练”方法，将自然语言描述转化为可复用的神经函数。编译时，教师模型生成特定任务样本，用于训练一个小型适配器以驱动紧凑解释器。生成的函数无需依赖教师模型，可存储、版本化并像普通软件一样组合使用。在FuzzyBench-Hard（快速编译器无法获得精确匹配的子集）上，该方法实现83.6%的语义准确率。虽然编译耗时约一分钟（远高于快速编译器的秒级），但结果更优。作者已将该编译器部署为公开交互服务，并在多站点网站助手、语言控制3D虚拟人及双向英-克劳迪什翻译器中验证了其应用效果。
原文摘要 · Abstract (English)
Many recurring text functions are easy to describe but difficult to implement with rules, while calling a large remote model for every input introduces repeated cost, latency, and dependency on a provider. We present compile by training, which turns a natural-language specification into a reusable neural function. At compile time, teacher models generate task-specific examples that are used to train a small adapter for a compact interpreter. The resulting function runs without the teachers and can be stored, versioned, and composed like ordinary software. On FuzzyBench-Hard, a subset on which the Program-as-Weights fast compiler produced no exact matches, compile by training reaches 83.6% semantic accuracy. This higher accuracy comes with a higher compile-time cost: roughly a minute rather than seconds for the fast compiler. We deploy the compiler in a public interactive service and demonstrate compiled functions in a multi-site website helper, a language-controlled 3D avatar, and a bidirectional English-Claudish translator.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。