大语言模型工具内学习的可证明优势

摘要

配备检索、记忆或外部API的工具增强型语言模型正在重塑人工智能领域，然而其理论优势仍未得到充分探索。本文通过展示在工具内学习（外部检索）相较于权重内学习（记忆）在事实回忆方面的优势，来探讨这一问题。我们证明，模型仅凭其权重所能记忆的事实数量从根本上受限于其参数规模。相反，我们证实，通过一种简单高效的电路构建，工具使用能够实现无限制的事实回忆。这些结论在控制实验中得到了验证，其中使用工具的模型始终优于依赖记忆的模型。我们进一步表明，对于预训练的大型语言模型，教授工具使用和通用规则比将事实微调至记忆中更为有效。本研究不仅提供了理论基础，还通过实证确立了工具增强型工作流程不仅实用，而且在可扩展性上具有可证明的优势。

English

Tool-augmented language models, equipped with retrieval, memory, or external APIs, are reshaping AI, yet their theoretical advantages remain underexplored. In this paper, we address this question by demonstrating the benefits of in-tool learning (external retrieval) over in-weight learning (memorization) for factual recall. We show that the number of facts a model can memorize solely in its weights is fundamentally limited by its parameter count. In contrast, we prove that tool-use enables unbounded factual recall via a simple and efficient circuit construction. These results are validated in controlled experiments, where tool-using models consistently outperform memorizing ones. We further show that for pretrained large language models, teaching tool-use and general rules is more effective than finetuning facts into memory. Our work provides both a theoretical and empirical foundation, establishing why tool-augmented workflows are not just practical, but provably more scalable.

大语言模型工具内学习的可证明优势

Provable Benefits of In-Tool Learning for Large Language Models

摘要

Support