We address the zero-shot transfer learning setting for the knowledge base question answering (KBQA) problem, where a large volume of labeled training data is available for the source domain, but no such labeled examples are available for the target domain. Transfer learning for KBQA makes use of large volumes of unlabeled data in the target in addition to the labeled data in the source. More recently, few-shot in-context learning using Black-box Large Language Models (BLLMs) has been adapted for KBQA without considering any source domain data. In this work, we show how to meaningfully combine these two paradigms for KBQA so that their benefits add up. Specifically, we preserve the two stage retrieve-then-generate pipeline of supervised KBQA and introduce interaction between in-context learning using BLLMs and transfer learning from the source for both stages. In addition, we propose execution-guided self-refinement using BLLMs, decoupled from the transfer setting. With the help of experiments using benchmark datasets GrailQA as the source and WebQSP as the target, we show that the proposed combination brings significant improvements to both stages and also outperforms by a large margin state-of-the-art supervised KBQA models trained on the source. We also show that in the in-domain setting, the proposed BLLM augmentation significantly outperforms state-of-the-art supervised models, when the volume of labeled data is limited, and also outperforms these marginally even when using the entire large training dataset.

翻译：我们针对知识库问答（KBQA）中的零样本迁移学习设定展开研究，该设定中源领域拥有大量标注训练数据，但目标领域缺乏此类标注样本。KBQA的迁移学习在利用源领域标注数据的同时，还使用了目标领域的大量无标注数据。近期，基于黑盒大语言模型（BLLMs）的少样本上下文学习方法已被应用于KBQA任务，但未考虑任何源领域数据。本研究展示了如何将这两种范式有效结合以提升KBQA性能，从而实现优势互补。具体而言，我们保留了监督式KBQA的两阶段"检索-生成"流程，并在两个阶段中引入了基于BLLMs的上下文学习与源领域迁移学习之间的交互。此外，我们提出了与迁移设置解耦的、基于BLLMs的执行引导自优化方法。通过以GrailQA为源领域、WebQSP为目标领域的基准数据集实验，我们证明了所提出的组合方法在两个阶段均带来显著提升，且大幅优于在源领域训练的最先进监督式KBQA模型。在领域内设定下，当标注数据量有限时，所提出的BLLMs增强方法显著超越最先进监督模型；即便使用完整大规模训练数据集时，该方法仍能保持微弱优势。