Large Language Models (LLMs) can generate SQL queries from natural language questions but struggle with database-specific schemas and tacit domain knowledge. We introduce a framework for continual learning from human feedback in text-to-SQL, where a learning agent receives natural language feedback to refine queries and distills the revealed knowledge for reuse on future tasks. This distilled knowledge is stored in a structured memory, enabling the agent to improve execution accuracy over time. We design and evaluate multiple variations of a learning agent architecture that vary in how they capture and retrieve past experiences. Experiments on the BIRD benchmark Dev set show that memory-augmented agents, particularly the Procedural Agent, achieve significant accuracy gains and error reduction by leveraging human-in-the-loop feedback. Our results highlight the importance of transforming tacit human expertise into reusable knowledge, paving the way for more adaptive, domain-aware text-to-SQL systems that continually learn from a human-in-the-loop.

翻译：大型语言模型（LLMs）能够根据自然语言问题生成SQL查询，但在处理特定数据库模式及隐含领域知识时仍面临挑战。本文提出一种基于人类反馈的文本到SQL持续学习框架：学习智能体接收自然语言反馈以优化查询，并通过提炼反馈中揭示的知识将其复用于后续任务。这些提炼后的知识存储于结构化记忆模块中，使智能体能够随时间提升执行准确率。我们设计并评估了多种学习智能体架构变体，重点比较其捕获与检索历史经验的不同机制。在BIRD基准开发集上的实验表明，记忆增强型智能体（特别是过程型智能体）通过利用人机交互反馈，实现了显著的准确率提升与错误率降低。研究结果凸显了将隐性人类专业知识转化为可复用知识的重要性，为构建更具适应性、领域感知能力且能持续从人机交互中学习的文本到SQL系统开辟了道路。